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.
235 prompt "Debug information"
236 depends on DEBUG_KERNEL
238 Selecting something other than "None" results in a kernel image
239 that will include debugging info resulting in a larger kernel image.
240 This adds debug symbols to the kernel and modules (gcc -g), and
241 is needed if you intend to use kernel crashdump or binary object
242 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
244 Choose which version of DWARF debug info to emit. If unsure,
245 select "Toolchain default".
247 config DEBUG_INFO_NONE
248 bool "Disable debug information"
250 Do not build the kernel with debugging information, which will
251 result in a faster and smaller build.
253 config DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT
254 bool "Rely on the toolchain's implicit default DWARF version"
257 The implicit default version of DWARF debug info produced by a
258 toolchain changes over time.
260 This can break consumers of the debug info that haven't upgraded to
261 support newer revisions, and prevent testing newer versions, but
262 those should be less common scenarios.
264 config DEBUG_INFO_DWARF4
265 bool "Generate DWARF Version 4 debuginfo"
267 depends on !CC_IS_CLANG || (CC_IS_CLANG && (AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502)))
269 Generate DWARF v4 debug info. This requires gcc 4.5+, binutils 2.35.2
270 if using clang without clang's integrated assembler, and gdb 7.0+.
272 If you have consumers of DWARF debug info that are not ready for
273 newer revisions of DWARF, you may wish to choose this or have your
276 config DEBUG_INFO_DWARF5
277 bool "Generate DWARF Version 5 debuginfo"
279 depends on !CC_IS_CLANG || (CC_IS_CLANG && (AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502)))
281 Generate DWARF v5 debug info. Requires binutils 2.35.2, gcc 5.0+ (gcc
282 5.0+ accepts the -gdwarf-5 flag but only had partial support for some
283 draft features until 7.0), and gdb 8.0+.
285 Changes to the structure of debug info in Version 5 allow for around
286 15-18% savings in resulting image and debug info section sizes as
287 compared to DWARF Version 4. DWARF Version 5 standardizes previous
288 extensions such as accelerators for symbol indexing and the format
289 for fission (.dwo/.dwp) files. Users may not want to select this
290 config if they rely on tooling that has not yet been updated to
291 support DWARF Version 5.
293 endchoice # "Debug information"
297 config DEBUG_INFO_REDUCED
298 bool "Reduce debugging information"
300 If you say Y here gcc is instructed to generate less debugging
301 information for structure types. This means that tools that
302 need full debugging information (like kgdb or systemtap) won't
303 be happy. But if you merely need debugging information to
304 resolve line numbers there is no loss. Advantage is that
305 build directory object sizes shrink dramatically over a full
306 DEBUG_INFO build and compile times are reduced too.
307 Only works with newer gcc versions.
309 config DEBUG_INFO_COMPRESSED
310 bool "Compressed debugging information"
311 depends on $(cc-option,-gz=zlib)
312 depends on $(ld-option,--compress-debug-sections=zlib)
314 Compress the debug information using zlib. Requires GCC 5.0+ or Clang
315 5.0+, binutils 2.26+, and zlib.
317 Users of dpkg-deb via scripts/package/builddeb may find an increase in
318 size of their debug .deb packages with this config set, due to the
319 debug info being compressed with zlib, then the object files being
320 recompressed with a different compression scheme. But this is still
321 preferable to setting $KDEB_COMPRESS to "none" which would be even
324 config DEBUG_INFO_SPLIT
325 bool "Produce split debuginfo in .dwo files"
326 depends on $(cc-option,-gsplit-dwarf)
328 Generate debug info into separate .dwo files. This significantly
329 reduces the build directory size for builds with DEBUG_INFO,
330 because it stores the information only once on disk in .dwo
331 files instead of multiple times in object files and executables.
332 In addition the debug information is also compressed.
334 Requires recent gcc (4.7+) and recent gdb/binutils.
335 Any tool that packages or reads debug information would need
336 to know about the .dwo files and include them.
337 Incompatible with older versions of ccache.
339 config DEBUG_INFO_BTF
340 bool "Generate BTF typeinfo"
341 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED
342 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST
343 depends on BPF_SYSCALL
344 depends on !DEBUG_INFO_DWARF5 || PAHOLE_VERSION >= 121
346 Generate deduplicated BTF type information from DWARF debug info.
347 Turning this on expects presence of pahole tool, which will convert
348 DWARF type info into equivalent deduplicated BTF type info.
350 config PAHOLE_HAS_SPLIT_BTF
351 def_bool PAHOLE_VERSION >= 119
353 config PAHOLE_HAS_BTF_TAG
354 def_bool PAHOLE_VERSION >= 123
355 depends on CC_IS_CLANG
357 Decide whether pahole emits btf_tag attributes (btf_type_tag and
358 btf_decl_tag) or not. Currently only clang compiler implements
359 these attributes, so make the config depend on CC_IS_CLANG.
361 config DEBUG_INFO_BTF_MODULES
363 depends on DEBUG_INFO_BTF && MODULES && PAHOLE_HAS_SPLIT_BTF
365 Generate compact split BTF type information for kernel modules.
367 config MODULE_ALLOW_BTF_MISMATCH
368 bool "Allow loading modules with non-matching BTF type info"
369 depends on DEBUG_INFO_BTF_MODULES
371 For modules whose split BTF does not match vmlinux, load without
372 BTF rather than refusing to load. The default behavior with
373 module BTF enabled is to reject modules with such mismatches;
374 this option will still load module BTF where possible but ignore
375 it when a mismatch is found.
378 bool "Provide GDB scripts for kernel debugging"
380 This creates the required links to GDB helper scripts in the
381 build directory. If you load vmlinux into gdb, the helper
382 scripts will be automatically imported by gdb as well, and
383 additional functions are available to analyze a Linux kernel
384 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst
390 int "Warn for stack frames larger than"
392 default 2048 if GCC_PLUGIN_LATENT_ENTROPY
393 default 2048 if PARISC
394 default 1536 if (!64BIT && XTENSA)
395 default 1024 if !64BIT
396 default 2048 if 64BIT
398 Tell gcc to warn at build time for stack frames larger than this.
399 Setting this too low will cause a lot of warnings.
400 Setting it to 0 disables the warning.
402 config STRIP_ASM_SYMS
403 bool "Strip assembler-generated symbols during link"
406 Strip internal assembler-generated symbols during a link (symbols
407 that look like '.Lxxx') so they don't pollute the output of
408 get_wchan() and suchlike.
411 bool "Generate readable assembler code"
412 depends on DEBUG_KERNEL
415 Disable some compiler optimizations that tend to generate human unreadable
416 assembler output. This may make the kernel slightly slower, but it helps
417 to keep kernel developers who have to stare a lot at assembler listings
420 config HEADERS_INSTALL
421 bool "Install uapi headers to usr/include"
424 This option will install uapi headers (headers exported to user-space)
425 into the usr/include directory for use during the kernel build.
426 This is unneeded for building the kernel itself, but needed for some
427 user-space program samples. It is also needed by some features such
428 as uapi header sanity checks.
430 config DEBUG_SECTION_MISMATCH
431 bool "Enable full Section mismatch analysis"
434 The section mismatch analysis checks if there are illegal
435 references from one section to another section.
436 During linktime or runtime, some sections are dropped;
437 any use of code/data previously in these sections would
438 most likely result in an oops.
439 In the code, functions and variables are annotated with
440 __init,, etc. (see the full list in include/linux/init.h),
441 which results in the code/data being placed in specific sections.
442 The section mismatch analysis is always performed after a full
443 kernel build, and enabling this option causes the following
444 additional step to occur:
445 - Add the option -fno-inline-functions-called-once to gcc commands.
446 When inlining a function annotated with __init in a non-init
447 function, we would lose the section information and thus
448 the analysis would not catch the illegal reference.
449 This option tells gcc to inline less (but it does result in
452 config SECTION_MISMATCH_WARN_ONLY
453 bool "Make section mismatch errors non-fatal"
456 If you say N here, the build process will fail if there are any
457 section mismatch, instead of just throwing warnings.
461 config DEBUG_FORCE_FUNCTION_ALIGN_64B
462 bool "Force all function address 64B aligned"
463 depends on EXPERT && (X86_64 || ARM64 || PPC32 || PPC64 || ARC)
465 There are cases that a commit from one domain changes the function
466 address alignment of other domains, and cause magic performance
467 bump (regression or improvement). Enable this option will help to
468 verify if the bump is caused by function alignment changes, while
469 it will slightly increase the kernel size and affect icache usage.
471 It is mainly for debug and performance tuning use.
474 # Select this config option from the architecture Kconfig, if it
475 # is preferred to always offer frame pointers as a config
476 # option on the architecture (regardless of KERNEL_DEBUG):
478 config ARCH_WANT_FRAME_POINTERS
482 bool "Compile the kernel with frame pointers"
483 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS
484 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
486 If you say Y here the resulting kernel image will be slightly
487 larger and slower, but it gives very useful debugging information
488 in case of kernel bugs. (precise oopses/stacktraces/warnings)
493 config STACK_VALIDATION
494 bool "Compile-time stack metadata validation"
495 depends on HAVE_STACK_VALIDATION && UNWINDER_FRAME_POINTER
499 Validate frame pointer rules at compile-time. This helps ensure that
500 runtime stack traces are more reliable.
502 For more information, see
503 tools/objtool/Documentation/objtool.txt.
505 config NOINSTR_VALIDATION
507 depends on HAVE_NOINSTR_VALIDATION && DEBUG_ENTRY
512 bool "Generate vmlinux.map file when linking"
515 Selecting this option will pass "-Map=vmlinux.map" to ld
516 when linking vmlinux. That file can be useful for verifying
517 and debugging magic section games, and for seeing which
518 pieces of code get eliminated with
519 CONFIG_LD_DEAD_CODE_DATA_ELIMINATION.
521 config DEBUG_FORCE_WEAK_PER_CPU
522 bool "Force weak per-cpu definitions"
523 depends on DEBUG_KERNEL
525 s390 and alpha require percpu variables in modules to be
526 defined weak to work around addressing range issue which
527 puts the following two restrictions on percpu variable
530 1. percpu symbols must be unique whether static or not
531 2. percpu variables can't be defined inside a function
533 To ensure that generic code follows the above rules, this
534 option forces all percpu variables to be defined as weak.
536 endmenu # "Compiler options"
538 menu "Generic Kernel Debugging Instruments"
541 bool "Magic SysRq key"
544 If you say Y here, you will have some control over the system even
545 if the system crashes for example during kernel debugging (e.g., you
546 will be able to flush the buffer cache to disk, reboot the system
547 immediately or dump some status information). This is accomplished
548 by pressing various keys while holding SysRq (Alt+PrintScreen). It
549 also works on a serial console (on PC hardware at least), if you
550 send a BREAK and then within 5 seconds a command keypress. The
551 keys are documented in <file:Documentation/admin-guide/sysrq.rst>.
552 Don't say Y unless you really know what this hack does.
554 config MAGIC_SYSRQ_DEFAULT_ENABLE
555 hex "Enable magic SysRq key functions by default"
556 depends on MAGIC_SYSRQ
559 Specifies which SysRq key functions are enabled by default.
560 This may be set to 1 or 0 to enable or disable them all, or
561 to a bitmask as described in Documentation/admin-guide/sysrq.rst.
563 config MAGIC_SYSRQ_SERIAL
564 bool "Enable magic SysRq key over serial"
565 depends on MAGIC_SYSRQ
568 Many embedded boards have a disconnected TTL level serial which can
569 generate some garbage that can lead to spurious false sysrq detects.
570 This option allows you to decide whether you want to enable the
573 config MAGIC_SYSRQ_SERIAL_SEQUENCE
574 string "Char sequence that enables magic SysRq over serial"
575 depends on MAGIC_SYSRQ_SERIAL
578 Specifies a sequence of characters that can follow BREAK to enable
579 SysRq on a serial console.
581 If unsure, leave an empty string and the option will not be enabled.
584 bool "Debug Filesystem"
586 debugfs is a virtual file system that kernel developers use to put
587 debugging files into. Enable this option to be able to read and
588 write to these files.
590 For detailed documentation on the debugfs API, see
591 Documentation/filesystems/.
596 prompt "Debugfs default access"
598 default DEBUG_FS_ALLOW_ALL
600 This selects the default access restrictions for debugfs.
601 It can be overridden with kernel command line option
602 debugfs=[on,no-mount,off]. The restrictions apply for API access
603 and filesystem registration.
605 config DEBUG_FS_ALLOW_ALL
608 No restrictions apply. Both API and filesystem registration
609 is on. This is the normal default operation.
611 config DEBUG_FS_DISALLOW_MOUNT
612 bool "Do not register debugfs as filesystem"
614 The API is open but filesystem is not loaded. Clients can still do
615 their work and read with debug tools that do not need
618 config DEBUG_FS_ALLOW_NONE
621 Access is off. Clients get -PERM when trying to create nodes in
622 debugfs tree and debugfs is not registered as a filesystem.
623 Client can then back-off or continue without debugfs access.
627 source "lib/Kconfig.kgdb"
628 source "lib/Kconfig.ubsan"
629 source "lib/Kconfig.kcsan"
633 menu "Networking Debugging"
635 source "net/Kconfig.debug"
637 endmenu # "Networking Debugging"
639 menu "Memory Debugging"
641 source "mm/Kconfig.debug"
644 bool "Debug object operations"
645 depends on DEBUG_KERNEL
647 If you say Y here, additional code will be inserted into the
648 kernel to track the life time of various objects and validate
649 the operations on those objects.
651 config DEBUG_OBJECTS_SELFTEST
652 bool "Debug objects selftest"
653 depends on DEBUG_OBJECTS
655 This enables the selftest of the object debug code.
657 config DEBUG_OBJECTS_FREE
658 bool "Debug objects in freed memory"
659 depends on DEBUG_OBJECTS
661 This enables checks whether a k/v free operation frees an area
662 which contains an object which has not been deactivated
663 properly. This can make kmalloc/kfree-intensive workloads
666 config DEBUG_OBJECTS_TIMERS
667 bool "Debug timer objects"
668 depends on DEBUG_OBJECTS
670 If you say Y here, additional code will be inserted into the
671 timer routines to track the life time of timer objects and
672 validate the timer operations.
674 config DEBUG_OBJECTS_WORK
675 bool "Debug work objects"
676 depends on DEBUG_OBJECTS
678 If you say Y here, additional code will be inserted into the
679 work queue routines to track the life time of work objects and
680 validate the work operations.
682 config DEBUG_OBJECTS_RCU_HEAD
683 bool "Debug RCU callbacks objects"
684 depends on DEBUG_OBJECTS
686 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
688 config DEBUG_OBJECTS_PERCPU_COUNTER
689 bool "Debug percpu counter objects"
690 depends on DEBUG_OBJECTS
692 If you say Y here, additional code will be inserted into the
693 percpu counter routines to track the life time of percpu counter
694 objects and validate the percpu counter operations.
696 config DEBUG_OBJECTS_ENABLE_DEFAULT
697 int "debug_objects bootup default value (0-1)"
700 depends on DEBUG_OBJECTS
702 Debug objects boot parameter default value
704 config SHRINKER_DEBUG
705 bool "Enable shrinker debugging support"
708 Say Y to enable the shrinker debugfs interface which provides
709 visibility into the kernel memory shrinkers subsystem.
710 Disable it to avoid an extra memory footprint.
712 config HAVE_DEBUG_KMEMLEAK
715 config DEBUG_KMEMLEAK
716 bool "Kernel memory leak detector"
717 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
719 select STACKTRACE if STACKTRACE_SUPPORT
723 Say Y here if you want to enable the memory leak
724 detector. The memory allocation/freeing is traced in a way
725 similar to the Boehm's conservative garbage collector, the
726 difference being that the orphan objects are not freed but
727 only shown in /sys/kernel/debug/kmemleak. Enabling this
728 feature will introduce an overhead to memory
729 allocations. See Documentation/dev-tools/kmemleak.rst for more
732 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
733 of finding leaks due to the slab objects poisoning.
735 In order to access the kmemleak file, debugfs needs to be
736 mounted (usually at /sys/kernel/debug).
738 config DEBUG_KMEMLEAK_MEM_POOL_SIZE
739 int "Kmemleak memory pool size"
740 depends on DEBUG_KMEMLEAK
744 Kmemleak must track all the memory allocations to avoid
745 reporting false positives. Since memory may be allocated or
746 freed before kmemleak is fully initialised, use a static pool
747 of metadata objects to track such callbacks. After kmemleak is
748 fully initialised, this memory pool acts as an emergency one
749 if slab allocations fail.
751 config DEBUG_KMEMLEAK_TEST
752 tristate "Simple test for the kernel memory leak detector"
753 depends on DEBUG_KMEMLEAK && m
755 This option enables a module that explicitly leaks memory.
759 config DEBUG_KMEMLEAK_DEFAULT_OFF
760 bool "Default kmemleak to off"
761 depends on DEBUG_KMEMLEAK
763 Say Y here to disable kmemleak by default. It can then be enabled
764 on the command line via kmemleak=on.
766 config DEBUG_KMEMLEAK_AUTO_SCAN
767 bool "Enable kmemleak auto scan thread on boot up"
769 depends on DEBUG_KMEMLEAK
771 Depending on the cpu, kmemleak scan may be cpu intensive and can
772 stall user tasks at times. This option enables/disables automatic
773 kmemleak scan at boot up.
775 Say N here to disable kmemleak auto scan thread to stop automatic
776 scanning. Disabling this option disables automatic reporting of
781 config DEBUG_STACK_USAGE
782 bool "Stack utilization instrumentation"
783 depends on DEBUG_KERNEL && !IA64
785 Enables the display of the minimum amount of free stack which each
786 task has ever had available in the sysrq-T and sysrq-P debug output.
788 This option will slow down process creation somewhat.
790 config SCHED_STACK_END_CHECK
791 bool "Detect stack corruption on calls to schedule()"
792 depends on DEBUG_KERNEL
795 This option checks for a stack overrun on calls to schedule().
796 If the stack end location is found to be over written always panic as
797 the content of the corrupted region can no longer be trusted.
798 This is to ensure no erroneous behaviour occurs which could result in
799 data corruption or a sporadic crash at a later stage once the region
800 is examined. The runtime overhead introduced is minimal.
802 config ARCH_HAS_DEBUG_VM_PGTABLE
805 An architecture should select this when it can successfully
806 build and run DEBUG_VM_PGTABLE.
808 config DEBUG_VM_IRQSOFF
809 def_bool DEBUG_VM && !PREEMPT_RT
813 depends on DEBUG_KERNEL
815 Enable this to turn on extended checks in the virtual-memory system
816 that may impact performance.
820 config DEBUG_VM_VMACACHE
821 bool "Debug VMA caching"
824 Enable this to turn on VMA caching debug information. Doing so
825 can cause significant overhead, so only enable it in non-production
831 bool "Debug VM red-black trees"
834 Enable VM red-black tree debugging information and extra validations.
838 config DEBUG_VM_PGFLAGS
839 bool "Debug page-flags operations"
842 Enables extra validation on page flags operations.
846 config DEBUG_VM_PGTABLE
847 bool "Debug arch page table for semantics compliance"
849 depends on ARCH_HAS_DEBUG_VM_PGTABLE
850 default y if DEBUG_VM
852 This option provides a debug method which can be used to test
853 architecture page table helper functions on various platforms in
854 verifying if they comply with expected generic MM semantics. This
855 will help architecture code in making sure that any changes or
856 new additions of these helpers still conform to expected
857 semantics of the generic MM. Platforms will have to opt in for
858 this through ARCH_HAS_DEBUG_VM_PGTABLE.
862 config ARCH_HAS_DEBUG_VIRTUAL
866 bool "Debug VM translations"
867 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL
869 Enable some costly sanity checks in virtual to page code. This can
870 catch mistakes with virt_to_page() and friends.
874 config DEBUG_NOMMU_REGIONS
875 bool "Debug the global anon/private NOMMU mapping region tree"
876 depends on DEBUG_KERNEL && !MMU
878 This option causes the global tree of anonymous and private mapping
879 regions to be regularly checked for invalid topology.
881 config DEBUG_MEMORY_INIT
882 bool "Debug memory initialisation" if EXPERT
885 Enable this for additional checks during memory initialisation.
886 The sanity checks verify aspects of the VM such as the memory model
887 and other information provided by the architecture. Verbose
888 information will be printed at KERN_DEBUG loglevel depending
889 on the mminit_loglevel= command-line option.
893 config MEMORY_NOTIFIER_ERROR_INJECT
894 tristate "Memory hotplug notifier error injection module"
895 depends on MEMORY_HOTPLUG && NOTIFIER_ERROR_INJECTION
897 This option provides the ability to inject artificial errors to
898 memory hotplug notifier chain callbacks. It is controlled through
899 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
901 If the notifier call chain should be failed with some events
902 notified, write the error code to "actions/<notifier event>/error".
904 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
906 # cd /sys/kernel/debug/notifier-error-inject/memory
907 # echo -12 > actions/MEM_GOING_OFFLINE/error
908 # echo offline > /sys/devices/system/memory/memoryXXX/state
909 bash: echo: write error: Cannot allocate memory
911 To compile this code as a module, choose M here: the module will
912 be called memory-notifier-error-inject.
916 config DEBUG_PER_CPU_MAPS
917 bool "Debug access to per_cpu maps"
918 depends on DEBUG_KERNEL
921 Say Y to verify that the per_cpu map being accessed has
922 been set up. This adds a fair amount of code to kernel memory
923 and decreases performance.
927 config DEBUG_KMAP_LOCAL
928 bool "Debug kmap_local temporary mappings"
929 depends on DEBUG_KERNEL && KMAP_LOCAL
931 This option enables additional error checking for the kmap_local
932 infrastructure. Disable for production use.
934 config ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
937 config DEBUG_KMAP_LOCAL_FORCE_MAP
938 bool "Enforce kmap_local temporary mappings"
939 depends on DEBUG_KERNEL && ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
941 select DEBUG_KMAP_LOCAL
943 This option enforces temporary mappings through the kmap_local
944 mechanism for non-highmem pages and on non-highmem systems.
945 Disable this for production systems!
948 bool "Highmem debugging"
949 depends on DEBUG_KERNEL && HIGHMEM
950 select DEBUG_KMAP_LOCAL_FORCE_MAP if ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
951 select DEBUG_KMAP_LOCAL
953 This option enables additional error checking for high memory
954 systems. Disable for production systems.
956 config HAVE_DEBUG_STACKOVERFLOW
959 config DEBUG_STACKOVERFLOW
960 bool "Check for stack overflows"
961 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
963 Say Y here if you want to check for overflows of kernel, IRQ
964 and exception stacks (if your architecture uses them). This
965 option will show detailed messages if free stack space drops
966 below a certain limit.
968 These kinds of bugs usually occur when call-chains in the
969 kernel get too deep, especially when interrupts are
972 Use this in cases where you see apparently random memory
973 corruption, especially if it appears in 'struct thread_info'
975 If in doubt, say "N".
977 source "lib/Kconfig.kasan"
978 source "lib/Kconfig.kfence"
980 endmenu # "Memory Debugging"
983 bool "Debug shared IRQ handlers"
984 depends on DEBUG_KERNEL
986 Enable this to generate a spurious interrupt just before a shared
987 interrupt handler is deregistered (generating one when registering
988 is currently disabled). Drivers need to handle this correctly. Some
989 don't and need to be caught.
991 menu "Debug Oops, Lockups and Hangs"
996 Say Y here to enable the kernel to panic when it oopses. This
997 has the same effect as setting oops=panic on the kernel command
1000 This feature is useful to ensure that the kernel does not do
1001 anything erroneous after an oops which could result in data
1002 corruption or other issues.
1006 config PANIC_ON_OOPS_VALUE
1009 default 0 if !PANIC_ON_OOPS
1010 default 1 if PANIC_ON_OOPS
1012 config PANIC_TIMEOUT
1016 Set the timeout value (in seconds) until a reboot occurs when
1017 the kernel panics. If n = 0, then we wait forever. A timeout
1018 value n > 0 will wait n seconds before rebooting, while a timeout
1019 value n < 0 will reboot immediately.
1021 config LOCKUP_DETECTOR
1024 config SOFTLOCKUP_DETECTOR
1025 bool "Detect Soft Lockups"
1026 depends on DEBUG_KERNEL && !S390
1027 select LOCKUP_DETECTOR
1029 Say Y here to enable the kernel to act as a watchdog to detect
1032 Softlockups are bugs that cause the kernel to loop in kernel
1033 mode for more than 20 seconds, without giving other tasks a
1034 chance to run. The current stack trace is displayed upon
1035 detection and the system will stay locked up.
1037 config BOOTPARAM_SOFTLOCKUP_PANIC
1038 bool "Panic (Reboot) On Soft Lockups"
1039 depends on SOFTLOCKUP_DETECTOR
1041 Say Y here to enable the kernel to panic on "soft lockups",
1042 which are bugs that cause the kernel to loop in kernel
1043 mode for more than 20 seconds (configurable using the watchdog_thresh
1044 sysctl), without giving other tasks a chance to run.
1046 The panic can be used in combination with panic_timeout,
1047 to cause the system to reboot automatically after a
1048 lockup has been detected. This feature is useful for
1049 high-availability systems that have uptime guarantees and
1050 where a lockup must be resolved ASAP.
1054 config HARDLOCKUP_DETECTOR_PERF
1056 select SOFTLOCKUP_DETECTOR
1059 # Enables a timestamp based low pass filter to compensate for perf based
1060 # hard lockup detection which runs too fast due to turbo modes.
1062 config HARDLOCKUP_CHECK_TIMESTAMP
1066 # arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard
1067 # lockup detector rather than the perf based detector.
1069 config HARDLOCKUP_DETECTOR
1070 bool "Detect Hard Lockups"
1071 depends on DEBUG_KERNEL && !S390
1072 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH
1073 select LOCKUP_DETECTOR
1074 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF
1076 Say Y here to enable the kernel to act as a watchdog to detect
1079 Hardlockups are bugs that cause the CPU to loop in kernel mode
1080 for more than 10 seconds, without letting other interrupts have a
1081 chance to run. The current stack trace is displayed upon detection
1082 and the system will stay locked up.
1084 config BOOTPARAM_HARDLOCKUP_PANIC
1085 bool "Panic (Reboot) On Hard Lockups"
1086 depends on HARDLOCKUP_DETECTOR
1088 Say Y here to enable the kernel to panic on "hard lockups",
1089 which are bugs that cause the kernel to loop in kernel
1090 mode with interrupts disabled for more than 10 seconds (configurable
1091 using the watchdog_thresh sysctl).
1095 config DETECT_HUNG_TASK
1096 bool "Detect Hung Tasks"
1097 depends on DEBUG_KERNEL
1098 default SOFTLOCKUP_DETECTOR
1100 Say Y here to enable the kernel to detect "hung tasks",
1101 which are bugs that cause the task to be stuck in
1102 uninterruptible "D" state indefinitely.
1104 When a hung task is detected, the kernel will print the
1105 current stack trace (which you should report), but the
1106 task will stay in uninterruptible state. If lockdep is
1107 enabled then all held locks will also be reported. This
1108 feature has negligible overhead.
1110 config DEFAULT_HUNG_TASK_TIMEOUT
1111 int "Default timeout for hung task detection (in seconds)"
1112 depends on DETECT_HUNG_TASK
1115 This option controls the default timeout (in seconds) used
1116 to determine when a task has become non-responsive and should
1119 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
1120 sysctl or by writing a value to
1121 /proc/sys/kernel/hung_task_timeout_secs.
1123 A timeout of 0 disables the check. The default is two minutes.
1124 Keeping the default should be fine in most cases.
1126 config BOOTPARAM_HUNG_TASK_PANIC
1127 bool "Panic (Reboot) On Hung Tasks"
1128 depends on DETECT_HUNG_TASK
1130 Say Y here to enable the kernel to panic on "hung tasks",
1131 which are bugs that cause the kernel to leave a task stuck
1132 in uninterruptible "D" state.
1134 The panic can be used in combination with panic_timeout,
1135 to cause the system to reboot automatically after a
1136 hung task has been detected. This feature is useful for
1137 high-availability systems that have uptime guarantees and
1138 where a hung tasks must be resolved ASAP.
1143 bool "Detect Workqueue Stalls"
1144 depends on DEBUG_KERNEL
1146 Say Y here to enable stall detection on workqueues. If a
1147 worker pool doesn't make forward progress on a pending work
1148 item for over a given amount of time, 30s by default, a
1149 warning message is printed along with dump of workqueue
1150 state. This can be configured through kernel parameter
1151 "workqueue.watchdog_thresh" and its sysfs counterpart.
1154 tristate "Test module to generate lockups"
1157 This builds the "test_lockup" module that helps to make sure
1158 that watchdogs and lockup detectors are working properly.
1160 Depending on module parameters it could emulate soft or hard
1161 lockup, "hung task", or locking arbitrary lock for a long time.
1162 Also it could generate series of lockups with cooling-down periods.
1166 endmenu # "Debug lockups and hangs"
1168 menu "Scheduler Debugging"
1171 bool "Collect scheduler debugging info"
1172 depends on DEBUG_KERNEL && PROC_FS
1175 If you say Y here, the /proc/sched_debug file will be provided
1176 that can help debug the scheduler. The runtime overhead of this
1184 bool "Collect scheduler statistics"
1185 depends on DEBUG_KERNEL && PROC_FS
1188 If you say Y here, additional code will be inserted into the
1189 scheduler and related routines to collect statistics about
1190 scheduler behavior and provide them in /proc/schedstat. These
1191 stats may be useful for both tuning and debugging the scheduler
1192 If you aren't debugging the scheduler or trying to tune a specific
1193 application, you can say N to avoid the very slight overhead
1198 config DEBUG_TIMEKEEPING
1199 bool "Enable extra timekeeping sanity checking"
1201 This option will enable additional timekeeping sanity checks
1202 which may be helpful when diagnosing issues where timekeeping
1203 problems are suspected.
1205 This may include checks in the timekeeping hotpaths, so this
1206 option may have a (very small) performance impact to some
1211 config DEBUG_PREEMPT
1212 bool "Debug preemptible kernel"
1213 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT
1216 If you say Y here then the kernel will use a debug variant of the
1217 commonly used smp_processor_id() function and will print warnings
1218 if kernel code uses it in a preemption-unsafe way. Also, the kernel
1219 will detect preemption count underflows.
1221 menu "Lock Debugging (spinlocks, mutexes, etc...)"
1223 config LOCK_DEBUGGING_SUPPORT
1225 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1228 config PROVE_LOCKING
1229 bool "Lock debugging: prove locking correctness"
1230 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1232 select DEBUG_SPINLOCK
1233 select DEBUG_MUTEXES if !PREEMPT_RT
1234 select DEBUG_RT_MUTEXES if RT_MUTEXES
1236 select DEBUG_WW_MUTEX_SLOWPATH
1237 select DEBUG_LOCK_ALLOC
1238 select PREEMPT_COUNT if !ARCH_NO_PREEMPT
1239 select TRACE_IRQFLAGS
1242 This feature enables the kernel to prove that all locking
1243 that occurs in the kernel runtime is mathematically
1244 correct: that under no circumstance could an arbitrary (and
1245 not yet triggered) combination of observed locking
1246 sequences (on an arbitrary number of CPUs, running an
1247 arbitrary number of tasks and interrupt contexts) cause a
1250 In short, this feature enables the kernel to report locking
1251 related deadlocks before they actually occur.
1253 The proof does not depend on how hard and complex a
1254 deadlock scenario would be to trigger: how many
1255 participant CPUs, tasks and irq-contexts would be needed
1256 for it to trigger. The proof also does not depend on
1257 timing: if a race and a resulting deadlock is possible
1258 theoretically (no matter how unlikely the race scenario
1259 is), it will be proven so and will immediately be
1260 reported by the kernel (once the event is observed that
1261 makes the deadlock theoretically possible).
1263 If a deadlock is impossible (i.e. the locking rules, as
1264 observed by the kernel, are mathematically correct), the
1265 kernel reports nothing.
1267 NOTE: this feature can also be enabled for rwlocks, mutexes
1268 and rwsems - in which case all dependencies between these
1269 different locking variants are observed and mapped too, and
1270 the proof of observed correctness is also maintained for an
1271 arbitrary combination of these separate locking variants.
1273 For more details, see Documentation/locking/lockdep-design.rst.
1275 config PROVE_RAW_LOCK_NESTING
1276 bool "Enable raw_spinlock - spinlock nesting checks"
1277 depends on PROVE_LOCKING
1280 Enable the raw_spinlock vs. spinlock nesting checks which ensure
1281 that the lock nesting rules for PREEMPT_RT enabled kernels are
1284 NOTE: There are known nesting problems. So if you enable this
1285 option expect lockdep splats until these problems have been fully
1286 addressed which is work in progress. This config switch allows to
1287 identify and analyze these problems. It will be removed and the
1288 check permanently enabled once the main issues have been fixed.
1290 If unsure, select N.
1293 bool "Lock usage statistics"
1294 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1296 select DEBUG_SPINLOCK
1297 select DEBUG_MUTEXES if !PREEMPT_RT
1298 select DEBUG_RT_MUTEXES if RT_MUTEXES
1299 select DEBUG_LOCK_ALLOC
1302 This feature enables tracking lock contention points
1304 For more details, see Documentation/locking/lockstat.rst
1306 This also enables lock events required by "perf lock",
1308 If you want to use "perf lock", you also need to turn on
1309 CONFIG_EVENT_TRACING.
1311 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1312 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1314 config DEBUG_RT_MUTEXES
1315 bool "RT Mutex debugging, deadlock detection"
1316 depends on DEBUG_KERNEL && RT_MUTEXES
1318 This allows rt mutex semantics violations and rt mutex related
1319 deadlocks (lockups) to be detected and reported automatically.
1321 config DEBUG_SPINLOCK
1322 bool "Spinlock and rw-lock debugging: basic checks"
1323 depends on DEBUG_KERNEL
1324 select UNINLINE_SPIN_UNLOCK
1326 Say Y here and build SMP to catch missing spinlock initialization
1327 and certain other kinds of spinlock errors commonly made. This is
1328 best used in conjunction with the NMI watchdog so that spinlock
1329 deadlocks are also debuggable.
1331 config DEBUG_MUTEXES
1332 bool "Mutex debugging: basic checks"
1333 depends on DEBUG_KERNEL && !PREEMPT_RT
1335 This feature allows mutex semantics violations to be detected and
1338 config DEBUG_WW_MUTEX_SLOWPATH
1339 bool "Wait/wound mutex debugging: Slowpath testing"
1340 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1341 select DEBUG_LOCK_ALLOC
1342 select DEBUG_SPINLOCK
1343 select DEBUG_MUTEXES if !PREEMPT_RT
1344 select DEBUG_RT_MUTEXES if PREEMPT_RT
1346 This feature enables slowpath testing for w/w mutex users by
1347 injecting additional -EDEADLK wound/backoff cases. Together with
1348 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1349 will test all possible w/w mutex interface abuse with the
1350 exception of simply not acquiring all the required locks.
1351 Note that this feature can introduce significant overhead, so
1352 it really should not be enabled in a production or distro kernel,
1353 even a debug kernel. If you are a driver writer, enable it. If
1354 you are a distro, do not.
1357 bool "RW Semaphore debugging: basic checks"
1358 depends on DEBUG_KERNEL
1360 This debugging feature allows mismatched rw semaphore locks
1361 and unlocks to be detected and reported.
1363 config DEBUG_LOCK_ALLOC
1364 bool "Lock debugging: detect incorrect freeing of live locks"
1365 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1366 select DEBUG_SPINLOCK
1367 select DEBUG_MUTEXES if !PREEMPT_RT
1368 select DEBUG_RT_MUTEXES if RT_MUTEXES
1371 This feature will check whether any held lock (spinlock, rwlock,
1372 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1373 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1374 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1375 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1376 held during task exit.
1380 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1385 config LOCKDEP_SMALL
1389 int "Bitsize for MAX_LOCKDEP_ENTRIES"
1390 depends on LOCKDEP && !LOCKDEP_SMALL
1394 Try increasing this value if you hit "BUG: MAX_LOCKDEP_ENTRIES too low!" message.
1396 config LOCKDEP_CHAINS_BITS
1397 int "Bitsize for MAX_LOCKDEP_CHAINS"
1398 depends on LOCKDEP && !LOCKDEP_SMALL
1402 Try increasing this value if you hit "BUG: MAX_LOCKDEP_CHAINS too low!" message.
1404 config LOCKDEP_STACK_TRACE_BITS
1405 int "Bitsize for MAX_STACK_TRACE_ENTRIES"
1406 depends on LOCKDEP && !LOCKDEP_SMALL
1410 Try increasing this value if you hit "BUG: MAX_STACK_TRACE_ENTRIES too low!" message.
1412 config LOCKDEP_STACK_TRACE_HASH_BITS
1413 int "Bitsize for STACK_TRACE_HASH_SIZE"
1414 depends on LOCKDEP && !LOCKDEP_SMALL
1418 Try increasing this value if you need large MAX_STACK_TRACE_ENTRIES.
1420 config LOCKDEP_CIRCULAR_QUEUE_BITS
1421 int "Bitsize for elements in circular_queue struct"
1426 Try increasing this value if you hit "lockdep bfs error:-1" warning due to __cq_enqueue() failure.
1428 config DEBUG_LOCKDEP
1429 bool "Lock dependency engine debugging"
1430 depends on DEBUG_KERNEL && LOCKDEP
1431 select DEBUG_IRQFLAGS
1433 If you say Y here, the lock dependency engine will do
1434 additional runtime checks to debug itself, at the price
1435 of more runtime overhead.
1437 config DEBUG_ATOMIC_SLEEP
1438 bool "Sleep inside atomic section checking"
1439 select PREEMPT_COUNT
1440 depends on DEBUG_KERNEL
1441 depends on !ARCH_NO_PREEMPT
1443 If you say Y here, various routines which may sleep will become very
1444 noisy if they are called inside atomic sections: when a spinlock is
1445 held, inside an rcu read side critical section, inside preempt disabled
1446 sections, inside an interrupt, etc...
1448 config DEBUG_LOCKING_API_SELFTESTS
1449 bool "Locking API boot-time self-tests"
1450 depends on DEBUG_KERNEL
1452 Say Y here if you want the kernel to run a short self-test during
1453 bootup. The self-test checks whether common types of locking bugs
1454 are detected by debugging mechanisms or not. (if you disable
1455 lock debugging then those bugs won't be detected of course.)
1456 The following locking APIs are covered: spinlocks, rwlocks,
1459 config LOCK_TORTURE_TEST
1460 tristate "torture tests for locking"
1461 depends on DEBUG_KERNEL
1464 This option provides a kernel module that runs torture tests
1465 on kernel locking primitives. The kernel module may be built
1466 after the fact on the running kernel to be tested, if desired.
1468 Say Y here if you want kernel locking-primitive torture tests
1469 to be built into the kernel.
1470 Say M if you want these torture tests to build as a module.
1471 Say N if you are unsure.
1473 config WW_MUTEX_SELFTEST
1474 tristate "Wait/wound mutex selftests"
1476 This option provides a kernel module that runs tests on the
1477 on the struct ww_mutex locking API.
1479 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction
1480 with this test harness.
1482 Say M if you want these self tests to build as a module.
1483 Say N if you are unsure.
1485 config SCF_TORTURE_TEST
1486 tristate "torture tests for smp_call_function*()"
1487 depends on DEBUG_KERNEL
1490 This option provides a kernel module that runs torture tests
1491 on the smp_call_function() family of primitives. The kernel
1492 module may be built after the fact on the running kernel to
1493 be tested, if desired.
1495 config CSD_LOCK_WAIT_DEBUG
1496 bool "Debugging for csd_lock_wait(), called from smp_call_function*()"
1497 depends on DEBUG_KERNEL
1501 This option enables debug prints when CPUs are slow to respond
1502 to the smp_call_function*() IPI wrappers. These debug prints
1503 include the IPI handler function currently executing (if any)
1504 and relevant stack traces.
1506 endmenu # lock debugging
1508 config TRACE_IRQFLAGS
1509 depends on TRACE_IRQFLAGS_SUPPORT
1512 Enables hooks to interrupt enabling and disabling for
1513 either tracing or lock debugging.
1515 config TRACE_IRQFLAGS_NMI
1517 depends on TRACE_IRQFLAGS
1518 depends on TRACE_IRQFLAGS_NMI_SUPPORT
1520 config DEBUG_IRQFLAGS
1521 bool "Debug IRQ flag manipulation"
1523 Enables checks for potentially unsafe enabling or disabling of
1524 interrupts, such as calling raw_local_irq_restore() when interrupts
1528 bool "Stack backtrace support"
1529 depends on STACKTRACE_SUPPORT
1531 This option causes the kernel to create a /proc/pid/stack for
1532 every process, showing its current stack trace.
1533 It is also used by various kernel debugging features that require
1534 stack trace generation.
1536 config WARN_ALL_UNSEEDED_RANDOM
1537 bool "Warn for all uses of unseeded randomness"
1540 Some parts of the kernel contain bugs relating to their use of
1541 cryptographically secure random numbers before it's actually possible
1542 to generate those numbers securely. This setting ensures that these
1543 flaws don't go unnoticed, by enabling a message, should this ever
1544 occur. This will allow people with obscure setups to know when things
1545 are going wrong, so that they might contact developers about fixing
1548 Unfortunately, on some models of some architectures getting
1549 a fully seeded CRNG is extremely difficult, and so this can
1550 result in dmesg getting spammed for a surprisingly long
1551 time. This is really bad from a security perspective, and
1552 so architecture maintainers really need to do what they can
1553 to get the CRNG seeded sooner after the system is booted.
1554 However, since users cannot do anything actionable to
1555 address this, by default this option is disabled.
1557 Say Y here if you want to receive warnings for all uses of
1558 unseeded randomness. This will be of use primarily for
1559 those developers interested in improving the security of
1560 Linux kernels running on their architecture (or
1563 config DEBUG_KOBJECT
1564 bool "kobject debugging"
1565 depends on DEBUG_KERNEL
1567 If you say Y here, some extra kobject debugging messages will be sent
1570 config DEBUG_KOBJECT_RELEASE
1571 bool "kobject release debugging"
1572 depends on DEBUG_OBJECTS_TIMERS
1574 kobjects are reference counted objects. This means that their
1575 last reference count put is not predictable, and the kobject can
1576 live on past the point at which a driver decides to drop its
1577 initial reference to the kobject gained on allocation. An
1578 example of this would be a struct device which has just been
1581 However, some buggy drivers assume that after such an operation,
1582 the memory backing the kobject can be immediately freed. This
1583 goes completely against the principles of a refcounted object.
1585 If you say Y here, the kernel will delay the release of kobjects
1586 on the last reference count to improve the visibility of this
1587 kind of kobject release bug.
1589 config HAVE_DEBUG_BUGVERBOSE
1592 menu "Debug kernel data structures"
1595 bool "Debug linked list manipulation"
1596 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION
1598 Enable this to turn on extended checks in the linked-list
1604 bool "Debug priority linked list manipulation"
1605 depends on DEBUG_KERNEL
1607 Enable this to turn on extended checks in the priority-ordered
1608 linked-list (plist) walking routines. This checks the entire
1609 list multiple times during each manipulation.
1614 bool "Debug SG table operations"
1615 depends on DEBUG_KERNEL
1617 Enable this to turn on checks on scatter-gather tables. This can
1618 help find problems with drivers that do not properly initialize
1623 config DEBUG_NOTIFIERS
1624 bool "Debug notifier call chains"
1625 depends on DEBUG_KERNEL
1627 Enable this to turn on sanity checking for notifier call chains.
1628 This is most useful for kernel developers to make sure that
1629 modules properly unregister themselves from notifier chains.
1630 This is a relatively cheap check but if you care about maximum
1633 config BUG_ON_DATA_CORRUPTION
1634 bool "Trigger a BUG when data corruption is detected"
1637 Select this option if the kernel should BUG when it encounters
1638 data corruption in kernel memory structures when they get checked
1645 config DEBUG_CREDENTIALS
1646 bool "Debug credential management"
1647 depends on DEBUG_KERNEL
1649 Enable this to turn on some debug checking for credential
1650 management. The additional code keeps track of the number of
1651 pointers from task_structs to any given cred struct, and checks to
1652 see that this number never exceeds the usage count of the cred
1655 Furthermore, if SELinux is enabled, this also checks that the
1656 security pointer in the cred struct is never seen to be invalid.
1660 source "kernel/rcu/Kconfig.debug"
1662 config DEBUG_WQ_FORCE_RR_CPU
1663 bool "Force round-robin CPU selection for unbound work items"
1664 depends on DEBUG_KERNEL
1667 Workqueue used to implicitly guarantee that work items queued
1668 without explicit CPU specified are put on the local CPU. This
1669 guarantee is no longer true and while local CPU is still
1670 preferred work items may be put on foreign CPUs. Kernel
1671 parameter "workqueue.debug_force_rr_cpu" is added to force
1672 round-robin CPU selection to flush out usages which depend on the
1673 now broken guarantee. This config option enables the debug
1674 feature by default. When enabled, memory and cache locality will
1677 config CPU_HOTPLUG_STATE_CONTROL
1678 bool "Enable CPU hotplug state control"
1679 depends on DEBUG_KERNEL
1680 depends on HOTPLUG_CPU
1683 Allows to write steps between "offline" and "online" to the CPUs
1684 sysfs target file so states can be stepped granular. This is a debug
1685 option for now as the hotplug machinery cannot be stopped and
1686 restarted at arbitrary points yet.
1688 Say N if your are unsure.
1691 bool "Latency measuring infrastructure"
1692 depends on DEBUG_KERNEL
1693 depends on STACKTRACE_SUPPORT
1695 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
1701 Enable this option if you want to use the LatencyTOP tool
1702 to find out which userspace is blocking on what kernel operations.
1704 source "kernel/trace/Kconfig"
1706 config PROVIDE_OHCI1394_DMA_INIT
1707 bool "Remote debugging over FireWire early on boot"
1708 depends on PCI && X86
1710 If you want to debug problems which hang or crash the kernel early
1711 on boot and the crashing machine has a FireWire port, you can use
1712 this feature to remotely access the memory of the crashed machine
1713 over FireWire. This employs remote DMA as part of the OHCI1394
1714 specification which is now the standard for FireWire controllers.
1716 With remote DMA, you can monitor the printk buffer remotely using
1717 firescope and access all memory below 4GB using fireproxy from gdb.
1718 Even controlling a kernel debugger is possible using remote DMA.
1722 If ohci1394_dma=early is used as boot parameter, it will initialize
1723 all OHCI1394 controllers which are found in the PCI config space.
1725 As all changes to the FireWire bus such as enabling and disabling
1726 devices cause a bus reset and thereby disable remote DMA for all
1727 devices, be sure to have the cable plugged and FireWire enabled on
1728 the debugging host before booting the debug target for debugging.
1730 This code (~1k) is freed after boot. By then, the firewire stack
1731 in charge of the OHCI-1394 controllers should be used instead.
1733 See Documentation/core-api/debugging-via-ohci1394.rst for more information.
1735 source "samples/Kconfig"
1737 config ARCH_HAS_DEVMEM_IS_ALLOWED
1740 config STRICT_DEVMEM
1741 bool "Filter access to /dev/mem"
1742 depends on MMU && DEVMEM
1743 depends on ARCH_HAS_DEVMEM_IS_ALLOWED || GENERIC_LIB_DEVMEM_IS_ALLOWED
1744 default y if PPC || X86 || ARM64
1746 If this option is disabled, you allow userspace (root) access to all
1747 of memory, including kernel and userspace memory. Accidental
1748 access to this is obviously disastrous, but specific access can
1749 be used by people debugging the kernel. Note that with PAT support
1750 enabled, even in this case there are restrictions on /dev/mem
1751 use due to the cache aliasing requirements.
1753 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
1754 file only allows userspace access to PCI space and the BIOS code and
1755 data regions. This is sufficient for dosemu and X and all common
1760 config IO_STRICT_DEVMEM
1761 bool "Filter I/O access to /dev/mem"
1762 depends on STRICT_DEVMEM
1764 If this option is disabled, you allow userspace (root) access to all
1765 io-memory regardless of whether a driver is actively using that
1766 range. Accidental access to this is obviously disastrous, but
1767 specific access can be used by people debugging kernel drivers.
1769 If this option is switched on, the /dev/mem file only allows
1770 userspace access to *idle* io-memory ranges (see /proc/iomem) This
1771 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
1772 if the driver using a given range cannot be disabled.
1776 menu "$(SRCARCH) Debugging"
1778 source "arch/$(SRCARCH)/Kconfig.debug"
1782 menu "Kernel Testing and Coverage"
1784 source "lib/kunit/Kconfig"
1786 config NOTIFIER_ERROR_INJECTION
1787 tristate "Notifier error injection"
1788 depends on DEBUG_KERNEL
1791 This option provides the ability to inject artificial errors to
1792 specified notifier chain callbacks. It is useful to test the error
1793 handling of notifier call chain failures.
1797 config PM_NOTIFIER_ERROR_INJECT
1798 tristate "PM notifier error injection module"
1799 depends on PM && NOTIFIER_ERROR_INJECTION
1800 default m if PM_DEBUG
1802 This option provides the ability to inject artificial errors to
1803 PM notifier chain callbacks. It is controlled through debugfs
1804 interface /sys/kernel/debug/notifier-error-inject/pm
1806 If the notifier call chain should be failed with some events
1807 notified, write the error code to "actions/<notifier event>/error".
1809 Example: Inject PM suspend error (-12 = -ENOMEM)
1811 # cd /sys/kernel/debug/notifier-error-inject/pm/
1812 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1813 # echo mem > /sys/power/state
1814 bash: echo: write error: Cannot allocate memory
1816 To compile this code as a module, choose M here: the module will
1817 be called pm-notifier-error-inject.
1821 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1822 tristate "OF reconfig notifier error injection module"
1823 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1825 This option provides the ability to inject artificial errors to
1826 OF reconfig notifier chain callbacks. It is controlled
1827 through debugfs interface under
1828 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1830 If the notifier call chain should be failed with some events
1831 notified, write the error code to "actions/<notifier event>/error".
1833 To compile this code as a module, choose M here: the module will
1834 be called of-reconfig-notifier-error-inject.
1838 config NETDEV_NOTIFIER_ERROR_INJECT
1839 tristate "Netdev notifier error injection module"
1840 depends on NET && NOTIFIER_ERROR_INJECTION
1842 This option provides the ability to inject artificial errors to
1843 netdevice notifier chain callbacks. It is controlled through debugfs
1844 interface /sys/kernel/debug/notifier-error-inject/netdev
1846 If the notifier call chain should be failed with some events
1847 notified, write the error code to "actions/<notifier event>/error".
1849 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1851 # cd /sys/kernel/debug/notifier-error-inject/netdev
1852 # echo -22 > actions/NETDEV_CHANGEMTU/error
1853 # ip link set eth0 mtu 1024
1854 RTNETLINK answers: Invalid argument
1856 To compile this code as a module, choose M here: the module will
1857 be called netdev-notifier-error-inject.
1861 config FUNCTION_ERROR_INJECTION
1863 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES
1865 config FAULT_INJECTION
1866 bool "Fault-injection framework"
1867 depends on DEBUG_KERNEL
1869 Provide fault-injection framework.
1870 For more details, see Documentation/fault-injection/.
1873 bool "Fault-injection capability for kmalloc"
1874 depends on FAULT_INJECTION
1875 depends on SLAB || SLUB
1877 Provide fault-injection capability for kmalloc.
1879 config FAIL_PAGE_ALLOC
1880 bool "Fault-injection capability for alloc_pages()"
1881 depends on FAULT_INJECTION
1883 Provide fault-injection capability for alloc_pages().
1885 config FAULT_INJECTION_USERCOPY
1886 bool "Fault injection capability for usercopy functions"
1887 depends on FAULT_INJECTION
1889 Provides fault-injection capability to inject failures
1890 in usercopy functions (copy_from_user(), get_user(), ...).
1892 config FAIL_MAKE_REQUEST
1893 bool "Fault-injection capability for disk IO"
1894 depends on FAULT_INJECTION && BLOCK
1896 Provide fault-injection capability for disk IO.
1898 config FAIL_IO_TIMEOUT
1899 bool "Fault-injection capability for faking disk interrupts"
1900 depends on FAULT_INJECTION && BLOCK
1902 Provide fault-injection capability on end IO handling. This
1903 will make the block layer "forget" an interrupt as configured,
1904 thus exercising the error handling.
1906 Only works with drivers that use the generic timeout handling,
1907 for others it won't do anything.
1910 bool "Fault-injection capability for futexes"
1912 depends on FAULT_INJECTION && FUTEX
1914 Provide fault-injection capability for futexes.
1916 config FAULT_INJECTION_DEBUG_FS
1917 bool "Debugfs entries for fault-injection capabilities"
1918 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1920 Enable configuration of fault-injection capabilities via debugfs.
1922 config FAIL_FUNCTION
1923 bool "Fault-injection capability for functions"
1924 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION
1926 Provide function-based fault-injection capability.
1927 This will allow you to override a specific function with a return
1928 with given return value. As a result, function caller will see
1929 an error value and have to handle it. This is useful to test the
1930 error handling in various subsystems.
1932 config FAIL_MMC_REQUEST
1933 bool "Fault-injection capability for MMC IO"
1934 depends on FAULT_INJECTION_DEBUG_FS && MMC
1936 Provide fault-injection capability for MMC IO.
1937 This will make the mmc core return data errors. This is
1938 useful to test the error handling in the mmc block device
1939 and to test how the mmc host driver handles retries from
1943 bool "Fault-injection capability for SunRPC"
1944 depends on FAULT_INJECTION_DEBUG_FS && SUNRPC_DEBUG
1946 Provide fault-injection capability for SunRPC and
1949 config FAULT_INJECTION_STACKTRACE_FILTER
1950 bool "stacktrace filter for fault-injection capabilities"
1951 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1954 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
1956 Provide stacktrace filter for fault-injection capabilities
1958 config ARCH_HAS_KCOV
1961 An architecture should select this when it can successfully
1962 build and run with CONFIG_KCOV. This typically requires
1963 disabling instrumentation for some early boot code.
1965 config CC_HAS_SANCOV_TRACE_PC
1966 def_bool $(cc-option,-fsanitize-coverage=trace-pc)
1970 bool "Code coverage for fuzzing"
1971 depends on ARCH_HAS_KCOV
1972 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS
1973 depends on !ARCH_WANTS_NO_INSTR || HAVE_NOINSTR_HACK || \
1974 GCC_VERSION >= 120000 || CLANG_VERSION >= 130000
1976 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC
1977 select OBJTOOL if HAVE_NOINSTR_HACK
1979 KCOV exposes kernel code coverage information in a form suitable
1980 for coverage-guided fuzzing (randomized testing).
1982 If RANDOMIZE_BASE is enabled, PC values will not be stable across
1983 different machines and across reboots. If you need stable PC values,
1984 disable RANDOMIZE_BASE.
1986 For more details, see Documentation/dev-tools/kcov.rst.
1988 config KCOV_ENABLE_COMPARISONS
1989 bool "Enable comparison operands collection by KCOV"
1991 depends on $(cc-option,-fsanitize-coverage=trace-cmp)
1993 KCOV also exposes operands of every comparison in the instrumented
1994 code along with operand sizes and PCs of the comparison instructions.
1995 These operands can be used by fuzzing engines to improve the quality
1996 of fuzzing coverage.
1998 config KCOV_INSTRUMENT_ALL
1999 bool "Instrument all code by default"
2003 If you are doing generic system call fuzzing (like e.g. syzkaller),
2004 then you will want to instrument the whole kernel and you should
2005 say y here. If you are doing more targeted fuzzing (like e.g.
2006 filesystem fuzzing with AFL) then you will want to enable coverage
2007 for more specific subsets of files, and should say n here.
2009 config KCOV_IRQ_AREA_SIZE
2010 hex "Size of interrupt coverage collection area in words"
2014 KCOV uses preallocated per-cpu areas to collect coverage from
2015 soft interrupts. This specifies the size of those areas in the
2016 number of unsigned long words.
2018 menuconfig RUNTIME_TESTING_MENU
2019 bool "Runtime Testing"
2022 if RUNTIME_TESTING_MENU
2025 tristate "Linux Kernel Dump Test Tool Module"
2028 This module enables testing of the different dumping mechanisms by
2029 inducing system failures at predefined crash points.
2030 If you don't need it: say N
2031 Choose M here to compile this code as a module. The module will be
2034 Documentation on how to use the module can be found in
2035 Documentation/fault-injection/provoke-crashes.rst
2037 config CPUMASK_KUNIT_TEST
2038 tristate "KUnit test for cpumask" if !KUNIT_ALL_TESTS
2040 default KUNIT_ALL_TESTS
2042 Enable to turn on cpumask tests, running at boot or module load time.
2044 For more information on KUnit and unit tests in general, please refer
2045 to the KUnit documentation in Documentation/dev-tools/kunit/.
2049 config TEST_LIST_SORT
2050 tristate "Linked list sorting test" if !KUNIT_ALL_TESTS
2052 default KUNIT_ALL_TESTS
2054 Enable this to turn on 'list_sort()' function test. This test is
2055 executed only once during system boot (so affects only boot time),
2056 or at module load time.
2060 config TEST_MIN_HEAP
2061 tristate "Min heap test"
2062 depends on DEBUG_KERNEL || m
2064 Enable this to turn on min heap function tests. This test is
2065 executed only once during system boot (so affects only boot time),
2066 or at module load time.
2071 tristate "Array-based sort test" if !KUNIT_ALL_TESTS
2073 default KUNIT_ALL_TESTS
2075 This option enables the self-test function of 'sort()' at boot,
2076 or at module load time.
2081 tristate "64bit/32bit division and modulo test"
2082 depends on DEBUG_KERNEL || m
2084 Enable this to turn on 'do_div()' function test. This test is
2085 executed only once during system boot (so affects only boot time),
2086 or at module load time.
2090 config KPROBES_SANITY_TEST
2091 tristate "Kprobes sanity tests" if !KUNIT_ALL_TESTS
2092 depends on DEBUG_KERNEL
2095 default KUNIT_ALL_TESTS
2097 This option provides for testing basic kprobes functionality on
2098 boot. Samples of kprobe and kretprobe are inserted and
2099 verified for functionality.
2101 Say N if you are unsure.
2103 config FPROBE_SANITY_TEST
2104 bool "Self test for fprobe"
2105 depends on DEBUG_KERNEL
2109 This option will enable testing the fprobe when the system boot.
2110 A series of tests are made to verify that the fprobe is functioning
2113 Say N if you are unsure.
2115 config BACKTRACE_SELF_TEST
2116 tristate "Self test for the backtrace code"
2117 depends on DEBUG_KERNEL
2119 This option provides a kernel module that can be used to test
2120 the kernel stack backtrace code. This option is not useful
2121 for distributions or general kernels, but only for kernel
2122 developers working on architecture code.
2124 Note that if you want to also test saved backtraces, you will
2125 have to enable STACKTRACE as well.
2127 Say N if you are unsure.
2129 config TEST_REF_TRACKER
2130 tristate "Self test for reference tracker"
2131 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
2134 This option provides a kernel module performing tests
2135 using reference tracker infrastructure.
2137 Say N if you are unsure.
2140 tristate "Red-Black tree test"
2141 depends on DEBUG_KERNEL
2143 A benchmark measuring the performance of the rbtree library.
2144 Also includes rbtree invariant checks.
2146 config REED_SOLOMON_TEST
2147 tristate "Reed-Solomon library test"
2148 depends on DEBUG_KERNEL || m
2150 select REED_SOLOMON_ENC16
2151 select REED_SOLOMON_DEC16
2153 This option enables the self-test function of rslib at boot,
2154 or at module load time.
2158 config INTERVAL_TREE_TEST
2159 tristate "Interval tree test"
2160 depends on DEBUG_KERNEL
2161 select INTERVAL_TREE
2163 A benchmark measuring the performance of the interval tree library
2166 tristate "Per cpu operations test"
2167 depends on m && DEBUG_KERNEL
2169 Enable this option to build test module which validates per-cpu
2174 config ATOMIC64_SELFTEST
2175 tristate "Perform an atomic64_t self-test"
2177 Enable this option to test the atomic64_t functions at boot or
2178 at module load time.
2182 config ASYNC_RAID6_TEST
2183 tristate "Self test for hardware accelerated raid6 recovery"
2184 depends on ASYNC_RAID6_RECOV
2187 This is a one-shot self test that permutes through the
2188 recovery of all the possible two disk failure scenarios for a
2189 N-disk array. Recovery is performed with the asynchronous
2190 raid6 recovery routines, and will optionally use an offload
2191 engine if one is available.
2196 tristate "Test functions located in the hexdump module at runtime"
2198 config STRING_SELFTEST
2199 tristate "Test string functions at runtime"
2201 config TEST_STRING_HELPERS
2202 tristate "Test functions located in the string_helpers module at runtime"
2205 tristate "Test strscpy*() family of functions at runtime"
2208 tristate "Test kstrto*() family of functions at runtime"
2211 tristate "Test printf() family of functions at runtime"
2214 tristate "Test scanf() family of functions at runtime"
2217 tristate "Test bitmap_*() family of functions at runtime"
2219 Enable this option to test the bitmap functions at boot.
2224 tristate "Test functions located in the uuid module at runtime"
2227 tristate "Test the XArray code at runtime"
2229 config TEST_RHASHTABLE
2230 tristate "Perform selftest on resizable hash table"
2232 Enable this option to test the rhashtable functions at boot.
2237 tristate "Perform selftest on siphash functions"
2239 Enable this option to test the kernel's siphash (<linux/siphash.h>) hash
2240 functions on boot (or module load).
2242 This is intended to help people writing architecture-specific
2243 optimized versions. If unsure, say N.
2246 tristate "Perform selftest on IDA functions"
2249 tristate "Perform selftest on priority array manager"
2252 Enable this option to test priority array manager on boot
2257 config TEST_IRQ_TIMINGS
2258 bool "IRQ timings selftest"
2259 depends on IRQ_TIMINGS
2261 Enable this option to test the irq timings code on boot.
2266 tristate "Test module loading with 'hello world' module"
2269 This builds the "test_module" module that emits "Hello, world"
2270 on printk when loaded. It is designed to be used for basic
2271 evaluation of the module loading subsystem (for example when
2272 validating module verification). It lacks any extra dependencies,
2273 and will not normally be loaded by the system unless explicitly
2279 tristate "Test module for compilation of bitops operations"
2282 This builds the "test_bitops" module that is much like the
2283 TEST_LKM module except that it does a basic exercise of the
2284 set/clear_bit macros and get_count_order/long to make sure there are
2285 no compiler warnings from C=1 sparse checker or -Wextra
2286 compilations. It has no dependencies and doesn't run or load unless
2287 explicitly requested by name. for example: modprobe test_bitops.
2292 tristate "Test module for stress/performance analysis of vmalloc allocator"
2297 This builds the "test_vmalloc" module that should be used for
2298 stress and performance analysis. So, any new change for vmalloc
2299 subsystem can be evaluated from performance and stability point
2304 config TEST_USER_COPY
2305 tristate "Test user/kernel boundary protections"
2308 This builds the "test_user_copy" module that runs sanity checks
2309 on the copy_to/from_user infrastructure, making sure basic
2310 user/kernel boundary testing is working. If it fails to load,
2311 a regression has been detected in the user/kernel memory boundary
2317 tristate "Test BPF filter functionality"
2320 This builds the "test_bpf" module that runs various test vectors
2321 against the BPF interpreter or BPF JIT compiler depending on the
2322 current setting. This is in particular useful for BPF JIT compiler
2323 development, but also to run regression tests against changes in
2324 the interpreter code. It also enables test stubs for eBPF maps and
2325 verifier used by user space verifier testsuite.
2329 config TEST_BLACKHOLE_DEV
2330 tristate "Test blackhole netdev functionality"
2333 This builds the "test_blackhole_dev" module that validates the
2334 data path through this blackhole netdev.
2338 config FIND_BIT_BENCHMARK
2339 tristate "Test find_bit functions"
2341 This builds the "test_find_bit" module that measure find_*_bit()
2342 functions performance.
2346 config TEST_FIRMWARE
2347 tristate "Test firmware loading via userspace interface"
2348 depends on FW_LOADER
2350 This builds the "test_firmware" module that creates a userspace
2351 interface for testing firmware loading. This can be used to
2352 control the triggering of firmware loading without needing an
2353 actual firmware-using device. The contents can be rechecked by
2359 tristate "sysctl test driver"
2360 depends on PROC_SYSCTL
2362 This builds the "test_sysctl" module. This driver enables to test the
2363 proc sysctl interfaces available to drivers safely without affecting
2364 production knobs which might alter system functionality.
2368 config BITFIELD_KUNIT
2369 tristate "KUnit test bitfield functions at runtime" if !KUNIT_ALL_TESTS
2371 default KUNIT_ALL_TESTS
2373 Enable this option to test the bitfield functions at boot.
2375 KUnit tests run during boot and output the results to the debug log
2376 in TAP format (http://testanything.org/). Only useful for kernel devs
2377 running the KUnit test harness, and not intended for inclusion into a
2380 For more information on KUnit and unit tests in general please refer
2381 to the KUnit documentation in Documentation/dev-tools/kunit/.
2385 config HASH_KUNIT_TEST
2386 tristate "KUnit Test for integer hash functions" if !KUNIT_ALL_TESTS
2388 default KUNIT_ALL_TESTS
2390 Enable this option to test the kernel's string (<linux/stringhash.h>), and
2391 integer (<linux/hash.h>) hash functions on boot.
2393 KUnit tests run during boot and output the results to the debug log
2394 in TAP format (https://testanything.org/). Only useful for kernel devs
2395 running the KUnit test harness, and not intended for inclusion into a
2398 For more information on KUnit and unit tests in general please refer
2399 to the KUnit documentation in Documentation/dev-tools/kunit/.
2401 This is intended to help people writing architecture-specific
2402 optimized versions. If unsure, say N.
2404 config RESOURCE_KUNIT_TEST
2405 tristate "KUnit test for resource API" if !KUNIT_ALL_TESTS
2407 default KUNIT_ALL_TESTS
2409 This builds the resource API unit test.
2410 Tests the logic of API provided by resource.c and ioport.h.
2411 For more information on KUnit and unit tests in general please refer
2412 to the KUnit documentation in Documentation/dev-tools/kunit/.
2416 config SYSCTL_KUNIT_TEST
2417 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS
2419 default KUNIT_ALL_TESTS
2421 This builds the proc sysctl unit test, which runs on boot.
2422 Tests the API contract and implementation correctness of sysctl.
2423 For more information on KUnit and unit tests in general please refer
2424 to the KUnit documentation in Documentation/dev-tools/kunit/.
2428 config LIST_KUNIT_TEST
2429 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS
2431 default KUNIT_ALL_TESTS
2433 This builds the linked list KUnit test suite.
2434 It tests that the API and basic functionality of the list_head type
2435 and associated macros.
2437 KUnit tests run during boot and output the results to the debug log
2438 in TAP format (https://testanything.org/). Only useful for kernel devs
2439 running the KUnit test harness, and not intended for inclusion into a
2442 For more information on KUnit and unit tests in general please refer
2443 to the KUnit documentation in Documentation/dev-tools/kunit/.
2447 config LINEAR_RANGES_TEST
2448 tristate "KUnit test for linear_ranges"
2450 select LINEAR_RANGES
2452 This builds the linear_ranges unit test, which runs on boot.
2453 Tests the linear_ranges logic correctness.
2454 For more information on KUnit and unit tests in general please refer
2455 to the KUnit documentation in Documentation/dev-tools/kunit/.
2459 config CMDLINE_KUNIT_TEST
2460 tristate "KUnit test for cmdline API" if !KUNIT_ALL_TESTS
2462 default KUNIT_ALL_TESTS
2464 This builds the cmdline API unit test.
2465 Tests the logic of API provided by cmdline.c.
2466 For more information on KUnit and unit tests in general please refer
2467 to the KUnit documentation in Documentation/dev-tools/kunit/.
2472 tristate "KUnit test for bits.h" if !KUNIT_ALL_TESTS
2474 default KUNIT_ALL_TESTS
2476 This builds the bits unit test.
2477 Tests the logic of macros defined in bits.h.
2478 For more information on KUnit and unit tests in general please refer
2479 to the KUnit documentation in Documentation/dev-tools/kunit/.
2483 config SLUB_KUNIT_TEST
2484 tristate "KUnit test for SLUB cache error detection" if !KUNIT_ALL_TESTS
2485 depends on SLUB_DEBUG && KUNIT
2486 default KUNIT_ALL_TESTS
2488 This builds SLUB allocator unit test.
2489 Tests SLUB cache debugging functionality.
2490 For more information on KUnit and unit tests in general please refer
2491 to the KUnit documentation in Documentation/dev-tools/kunit/.
2495 config RATIONAL_KUNIT_TEST
2496 tristate "KUnit test for rational.c" if !KUNIT_ALL_TESTS
2497 depends on KUNIT && RATIONAL
2498 default KUNIT_ALL_TESTS
2500 This builds the rational math unit test.
2501 For more information on KUnit and unit tests in general please refer
2502 to the KUnit documentation in Documentation/dev-tools/kunit/.
2506 config MEMCPY_KUNIT_TEST
2507 tristate "Test memcpy(), memmove(), and memset() functions at runtime" if !KUNIT_ALL_TESTS
2509 default KUNIT_ALL_TESTS
2511 Builds unit tests for memcpy(), memmove(), and memset() functions.
2512 For more information on KUnit and unit tests in general please refer
2513 to the KUnit documentation in Documentation/dev-tools/kunit/.
2517 config IS_SIGNED_TYPE_KUNIT_TEST
2518 tristate "Test is_signed_type() macro" if !KUNIT_ALL_TESTS
2520 default KUNIT_ALL_TESTS
2522 Builds unit tests for the is_signed_type() macro.
2524 For more information on KUnit and unit tests in general please refer
2525 to the KUnit documentation in Documentation/dev-tools/kunit/.
2529 config OVERFLOW_KUNIT_TEST
2530 tristate "Test check_*_overflow() functions at runtime" if !KUNIT_ALL_TESTS
2532 default KUNIT_ALL_TESTS
2534 Builds unit tests for the check_*_overflow(), size_*(), allocation, and
2537 For more information on KUnit and unit tests in general please refer
2538 to the KUnit documentation in Documentation/dev-tools/kunit/.
2542 config STACKINIT_KUNIT_TEST
2543 tristate "Test level of stack variable initialization" if !KUNIT_ALL_TESTS
2545 default KUNIT_ALL_TESTS
2547 Test if the kernel is zero-initializing stack variables and
2548 padding. Coverage is controlled by compiler flags,
2549 CONFIG_INIT_STACK_ALL_PATTERN, CONFIG_INIT_STACK_ALL_ZERO,
2550 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF,
2551 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL.
2553 config FORTIFY_KUNIT_TEST
2554 tristate "Test fortified str*() and mem*() function internals at runtime" if !KUNIT_ALL_TESTS
2555 depends on KUNIT && FORTIFY_SOURCE
2556 default KUNIT_ALL_TESTS
2558 Builds unit tests for checking internals of FORTIFY_SOURCE as used
2559 by the str*() and mem*() family of functions. For testing runtime
2560 traps of FORTIFY_SOURCE, see LKDTM's "FORTIFY_*" tests.
2562 config HW_BREAKPOINT_KUNIT_TEST
2563 bool "Test hw_breakpoint constraints accounting" if !KUNIT_ALL_TESTS
2564 depends on HAVE_HW_BREAKPOINT
2566 default KUNIT_ALL_TESTS
2568 Tests for hw_breakpoint constraints accounting.
2573 tristate "udelay test driver"
2575 This builds the "udelay_test" module that helps to make sure
2576 that udelay() is working properly.
2580 config TEST_STATIC_KEYS
2581 tristate "Test static keys"
2584 Test the static key interfaces.
2588 config TEST_DYNAMIC_DEBUG
2589 tristate "Test DYNAMIC_DEBUG"
2590 depends on DYNAMIC_DEBUG
2592 This module registers a tracer callback to count enabled
2593 pr_debugs in a 'do_debugging' function, then alters their
2594 enablements, calls the function, and compares counts.
2599 tristate "kmod stress tester"
2601 depends on NETDEVICES && NET_CORE && INET # for TUN
2603 depends on PAGE_SIZE_LESS_THAN_256KB # for BTRFS
2609 Test the kernel's module loading mechanism: kmod. kmod implements
2610 support to load modules using the Linux kernel's usermode helper.
2611 This test provides a series of tests against kmod.
2613 Although technically you can either build test_kmod as a module or
2614 into the kernel we disallow building it into the kernel since
2615 it stress tests request_module() and this will very likely cause
2616 some issues by taking over precious threads available from other
2617 module load requests, ultimately this could be fatal.
2621 tools/testing/selftests/kmod/kmod.sh --help
2625 config TEST_DEBUG_VIRTUAL
2626 tristate "Test CONFIG_DEBUG_VIRTUAL feature"
2627 depends on DEBUG_VIRTUAL
2629 Test the kernel's ability to detect incorrect calls to
2630 virt_to_phys() done against the non-linear part of the
2631 kernel's virtual address map.
2635 config TEST_MEMCAT_P
2636 tristate "Test memcat_p() helper function"
2638 Test the memcat_p() helper for correctly merging two
2639 pointer arrays together.
2643 config TEST_LIVEPATCH
2644 tristate "Test livepatching"
2646 depends on DYNAMIC_DEBUG
2647 depends on LIVEPATCH
2650 Test kernel livepatching features for correctness. The tests will
2651 load test modules that will be livepatched in various scenarios.
2653 To run all the livepatching tests:
2655 make -C tools/testing/selftests TARGETS=livepatch run_tests
2657 Alternatively, individual tests may be invoked:
2659 tools/testing/selftests/livepatch/test-callbacks.sh
2660 tools/testing/selftests/livepatch/test-livepatch.sh
2661 tools/testing/selftests/livepatch/test-shadow-vars.sh
2666 tristate "Perform selftest on object aggreration manager"
2670 Enable this option to test object aggregation manager on boot
2674 tristate "Test heap/page initialization"
2676 Test if the kernel is zero-initializing heap and page allocations.
2677 This can be useful to test init_on_alloc and init_on_free features.
2682 tristate "Test HMM (Heterogeneous Memory Management)"
2683 depends on TRANSPARENT_HUGEPAGE
2684 depends on DEVICE_PRIVATE
2688 This is a pseudo device driver solely for testing HMM.
2689 Say M here if you want to build the HMM test module.
2690 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests.
2694 config TEST_FREE_PAGES
2695 tristate "Test freeing pages"
2697 Test that a memory leak does not occur due to a race between
2698 freeing a block of pages and a speculative page reference.
2699 Loading this module is safe if your kernel has the bug fixed.
2700 If the bug is not fixed, it will leak gigabytes of memory and
2701 probably OOM your system.
2704 tristate "Test floating point operations in kernel space"
2705 depends on X86 && !KCOV_INSTRUMENT_ALL
2707 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu
2708 which will trigger a sequence of floating point operations. This is used
2709 for self-testing floating point control register setting in
2714 config TEST_CLOCKSOURCE_WATCHDOG
2715 tristate "Test clocksource watchdog in kernel space"
2716 depends on CLOCKSOURCE_WATCHDOG
2718 Enable this option to create a kernel module that will trigger
2719 a test of the clocksource watchdog. This module may be loaded
2720 via modprobe or insmod in which case it will run upon being
2721 loaded, or it may be built in, in which case it will run
2726 endif # RUNTIME_TESTING_MENU
2728 config ARCH_USE_MEMTEST
2731 An architecture should select this when it uses early_memtest()
2732 during boot process.
2736 depends on ARCH_USE_MEMTEST
2738 This option adds a kernel parameter 'memtest', which allows memtest
2739 to be set and executed.
2740 memtest=0, mean disabled; -- default
2741 memtest=1, mean do 1 test pattern;
2743 memtest=17, mean do 17 test patterns.
2744 If you are unsure how to answer this question, answer N.
2748 config HYPERV_TESTING
2749 bool "Microsoft Hyper-V driver testing"
2751 depends on HYPERV && DEBUG_FS
2753 Select this option to enable Hyper-V vmbus testing.
2755 endmenu # "Kernel Testing and Coverage"
2759 config RUST_DEBUG_ASSERTIONS
2760 bool "Debug assertions"
2763 Enables rustc's `-Cdebug-assertions` codegen option.
2765 This flag lets you turn `cfg(debug_assertions)` conditional
2766 compilation on or off. This can be used to enable extra debugging
2767 code in development but not in production. For example, it controls
2768 the behavior of the standard library's `debug_assert!` macro.
2770 Note that this will apply to all Rust code, including `core`.
2774 config RUST_OVERFLOW_CHECKS
2775 bool "Overflow checks"
2779 Enables rustc's `-Coverflow-checks` codegen option.
2781 This flag allows you to control the behavior of runtime integer
2782 overflow. When overflow-checks are enabled, a Rust panic will occur
2785 Note that this will apply to all Rust code, including `core`.
2791 source "Documentation/Kconfig"
2793 endmenu # Kernel hacking