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"
211 menu "Compile-time checks and compiler options"
214 bool "Compile the kernel with debug info"
215 depends on DEBUG_KERNEL && !COMPILE_TEST
217 If you say Y here the resulting kernel image will include
218 debugging info resulting in a larger kernel image.
219 This adds debug symbols to the kernel and modules (gcc -g), and
220 is needed if you intend to use kernel crashdump or binary object
221 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
222 Say Y here only if you plan to debug the kernel.
228 config DEBUG_INFO_REDUCED
229 bool "Reduce debugging information"
231 If you say Y here gcc is instructed to generate less debugging
232 information for structure types. This means that tools that
233 need full debugging information (like kgdb or systemtap) won't
234 be happy. But if you merely need debugging information to
235 resolve line numbers there is no loss. Advantage is that
236 build directory object sizes shrink dramatically over a full
237 DEBUG_INFO build and compile times are reduced too.
238 Only works with newer gcc versions.
240 config DEBUG_INFO_COMPRESSED
241 bool "Compressed debugging information"
242 depends on $(cc-option,-gz=zlib)
243 depends on $(ld-option,--compress-debug-sections=zlib)
245 Compress the debug information using zlib. Requires GCC 5.0+ or Clang
246 5.0+, binutils 2.26+, and zlib.
248 Users of dpkg-deb via scripts/package/builddeb may find an increase in
249 size of their debug .deb packages with this config set, due to the
250 debug info being compressed with zlib, then the object files being
251 recompressed with a different compression scheme. But this is still
252 preferable to setting $KDEB_COMPRESS to "none" which would be even
255 config DEBUG_INFO_SPLIT
256 bool "Produce split debuginfo in .dwo files"
257 depends on $(cc-option,-gsplit-dwarf)
259 Generate debug info into separate .dwo files. This significantly
260 reduces the build directory size for builds with DEBUG_INFO,
261 because it stores the information only once on disk in .dwo
262 files instead of multiple times in object files and executables.
263 In addition the debug information is also compressed.
265 Requires recent gcc (4.7+) and recent gdb/binutils.
266 Any tool that packages or reads debug information would need
267 to know about the .dwo files and include them.
268 Incompatible with older versions of ccache.
271 prompt "DWARF version"
273 Which version of DWARF debug info to emit.
275 config DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT
276 bool "Rely on the toolchain's implicit default DWARF version"
278 The implicit default version of DWARF debug info produced by a
279 toolchain changes over time.
281 This can break consumers of the debug info that haven't upgraded to
282 support newer revisions, and prevent testing newer versions, but
283 those should be less common scenarios.
287 config DEBUG_INFO_DWARF4
288 bool "Generate DWARF Version 4 debuginfo"
290 Generate DWARF v4 debug info. This requires gcc 4.5+ and gdb 7.0+.
292 If you have consumers of DWARF debug info that are not ready for
293 newer revisions of DWARF, you may wish to choose this or have your
296 config DEBUG_INFO_DWARF5
297 bool "Generate DWARF Version 5 debuginfo"
298 depends on !CC_IS_CLANG || (CC_IS_CLANG && (AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502)))
299 depends on !DEBUG_INFO_BTF || PAHOLE_VERSION >= 121
301 Generate DWARF v5 debug info. Requires binutils 2.35.2, gcc 5.0+ (gcc
302 5.0+ accepts the -gdwarf-5 flag but only had partial support for some
303 draft features until 7.0), and gdb 8.0+.
305 Changes to the structure of debug info in Version 5 allow for around
306 15-18% savings in resulting image and debug info section sizes as
307 compared to DWARF Version 4. DWARF Version 5 standardizes previous
308 extensions such as accelerators for symbol indexing and the format
309 for fission (.dwo/.dwp) files. Users may not want to select this
310 config if they rely on tooling that has not yet been updated to
311 support DWARF Version 5.
313 endchoice # "DWARF version"
315 config DEBUG_INFO_BTF
316 bool "Generate BTF typeinfo"
317 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED
318 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST
319 depends on BPF_SYSCALL
321 Generate deduplicated BTF type information from DWARF debug info.
322 Turning this on expects presence of pahole tool, which will convert
323 DWARF type info into equivalent deduplicated BTF type info.
325 config PAHOLE_HAS_SPLIT_BTF
326 def_bool PAHOLE_VERSION >= 119
328 config PAHOLE_HAS_BTF_TAG
329 def_bool PAHOLE_VERSION >= 123
330 depends on CC_IS_CLANG
332 Decide whether pahole emits btf_tag attributes (btf_type_tag and
333 btf_decl_tag) or not. Currently only clang compiler implements
334 these attributes, so make the config depend on CC_IS_CLANG.
336 config DEBUG_INFO_BTF_MODULES
338 depends on DEBUG_INFO_BTF && MODULES && PAHOLE_HAS_SPLIT_BTF
340 Generate compact split BTF type information for kernel modules.
342 config MODULE_ALLOW_BTF_MISMATCH
343 bool "Allow loading modules with non-matching BTF type info"
344 depends on DEBUG_INFO_BTF_MODULES
346 For modules whose split BTF does not match vmlinux, load without
347 BTF rather than refusing to load. The default behavior with
348 module BTF enabled is to reject modules with such mismatches;
349 this option will still load module BTF where possible but ignore
350 it when a mismatch is found.
353 bool "Provide GDB scripts for kernel debugging"
355 This creates the required links to GDB helper scripts in the
356 build directory. If you load vmlinux into gdb, the helper
357 scripts will be automatically imported by gdb as well, and
358 additional functions are available to analyze a Linux kernel
359 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst
365 int "Warn for stack frames larger than"
367 default 2048 if GCC_PLUGIN_LATENT_ENTROPY
368 default 2048 if PARISC
369 default 1536 if (!64BIT && XTENSA)
370 default 1024 if !64BIT
371 default 2048 if 64BIT
373 Tell gcc to warn at build time for stack frames larger than this.
374 Setting this too low will cause a lot of warnings.
375 Setting it to 0 disables the warning.
377 config STRIP_ASM_SYMS
378 bool "Strip assembler-generated symbols during link"
381 Strip internal assembler-generated symbols during a link (symbols
382 that look like '.Lxxx') so they don't pollute the output of
383 get_wchan() and suchlike.
386 bool "Generate readable assembler code"
387 depends on DEBUG_KERNEL
390 Disable some compiler optimizations that tend to generate human unreadable
391 assembler output. This may make the kernel slightly slower, but it helps
392 to keep kernel developers who have to stare a lot at assembler listings
395 config HEADERS_INSTALL
396 bool "Install uapi headers to usr/include"
399 This option will install uapi headers (headers exported to user-space)
400 into the usr/include directory for use during the kernel build.
401 This is unneeded for building the kernel itself, but needed for some
402 user-space program samples. It is also needed by some features such
403 as uapi header sanity checks.
405 config DEBUG_SECTION_MISMATCH
406 bool "Enable full Section mismatch analysis"
409 The section mismatch analysis checks if there are illegal
410 references from one section to another section.
411 During linktime or runtime, some sections are dropped;
412 any use of code/data previously in these sections would
413 most likely result in an oops.
414 In the code, functions and variables are annotated with
415 __init,, etc. (see the full list in include/linux/init.h),
416 which results in the code/data being placed in specific sections.
417 The section mismatch analysis is always performed after a full
418 kernel build, and enabling this option causes the following
419 additional step to occur:
420 - Add the option -fno-inline-functions-called-once to gcc commands.
421 When inlining a function annotated with __init in a non-init
422 function, we would lose the section information and thus
423 the analysis would not catch the illegal reference.
424 This option tells gcc to inline less (but it does result in
427 config SECTION_MISMATCH_WARN_ONLY
428 bool "Make section mismatch errors non-fatal"
431 If you say N here, the build process will fail if there are any
432 section mismatch, instead of just throwing warnings.
436 config DEBUG_FORCE_FUNCTION_ALIGN_64B
437 bool "Force all function address 64B aligned" if EXPERT
439 There are cases that a commit from one domain changes the function
440 address alignment of other domains, and cause magic performance
441 bump (regression or improvement). Enable this option will help to
442 verify if the bump is caused by function alignment changes, while
443 it will slightly increase the kernel size and affect icache usage.
445 It is mainly for debug and performance tuning use.
448 # Select this config option from the architecture Kconfig, if it
449 # is preferred to always offer frame pointers as a config
450 # option on the architecture (regardless of KERNEL_DEBUG):
452 config ARCH_WANT_FRAME_POINTERS
456 bool "Compile the kernel with frame pointers"
457 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS
458 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
460 If you say Y here the resulting kernel image will be slightly
461 larger and slower, but it gives very useful debugging information
462 in case of kernel bugs. (precise oopses/stacktraces/warnings)
464 config STACK_VALIDATION
465 bool "Compile-time stack metadata validation"
466 depends on HAVE_STACK_VALIDATION
469 Add compile-time checks to validate stack metadata, including frame
470 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure
471 that runtime stack traces are more reliable.
473 This is also a prerequisite for generation of ORC unwind data, which
474 is needed for CONFIG_UNWINDER_ORC.
476 For more information, see
477 tools/objtool/Documentation/stack-validation.txt.
479 config VMLINUX_VALIDATION
481 depends on STACK_VALIDATION && DEBUG_ENTRY
485 bool "Generate vmlinux.map file when linking"
488 Selecting this option will pass "-Map=vmlinux.map" to ld
489 when linking vmlinux. That file can be useful for verifying
490 and debugging magic section games, and for seeing which
491 pieces of code get eliminated with
492 CONFIG_LD_DEAD_CODE_DATA_ELIMINATION.
494 config DEBUG_FORCE_WEAK_PER_CPU
495 bool "Force weak per-cpu definitions"
496 depends on DEBUG_KERNEL
498 s390 and alpha require percpu variables in modules to be
499 defined weak to work around addressing range issue which
500 puts the following two restrictions on percpu variable
503 1. percpu symbols must be unique whether static or not
504 2. percpu variables can't be defined inside a function
506 To ensure that generic code follows the above rules, this
507 option forces all percpu variables to be defined as weak.
509 endmenu # "Compiler options"
511 menu "Generic Kernel Debugging Instruments"
514 bool "Magic SysRq key"
517 If you say Y here, you will have some control over the system even
518 if the system crashes for example during kernel debugging (e.g., you
519 will be able to flush the buffer cache to disk, reboot the system
520 immediately or dump some status information). This is accomplished
521 by pressing various keys while holding SysRq (Alt+PrintScreen). It
522 also works on a serial console (on PC hardware at least), if you
523 send a BREAK and then within 5 seconds a command keypress. The
524 keys are documented in <file:Documentation/admin-guide/sysrq.rst>.
525 Don't say Y unless you really know what this hack does.
527 config MAGIC_SYSRQ_DEFAULT_ENABLE
528 hex "Enable magic SysRq key functions by default"
529 depends on MAGIC_SYSRQ
532 Specifies which SysRq key functions are enabled by default.
533 This may be set to 1 or 0 to enable or disable them all, or
534 to a bitmask as described in Documentation/admin-guide/sysrq.rst.
536 config MAGIC_SYSRQ_SERIAL
537 bool "Enable magic SysRq key over serial"
538 depends on MAGIC_SYSRQ
541 Many embedded boards have a disconnected TTL level serial which can
542 generate some garbage that can lead to spurious false sysrq detects.
543 This option allows you to decide whether you want to enable the
546 config MAGIC_SYSRQ_SERIAL_SEQUENCE
547 string "Char sequence that enables magic SysRq over serial"
548 depends on MAGIC_SYSRQ_SERIAL
551 Specifies a sequence of characters that can follow BREAK to enable
552 SysRq on a serial console.
554 If unsure, leave an empty string and the option will not be enabled.
557 bool "Debug Filesystem"
559 debugfs is a virtual file system that kernel developers use to put
560 debugging files into. Enable this option to be able to read and
561 write to these files.
563 For detailed documentation on the debugfs API, see
564 Documentation/filesystems/.
569 prompt "Debugfs default access"
571 default DEBUG_FS_ALLOW_ALL
573 This selects the default access restrictions for debugfs.
574 It can be overridden with kernel command line option
575 debugfs=[on,no-mount,off]. The restrictions apply for API access
576 and filesystem registration.
578 config DEBUG_FS_ALLOW_ALL
581 No restrictions apply. Both API and filesystem registration
582 is on. This is the normal default operation.
584 config DEBUG_FS_DISALLOW_MOUNT
585 bool "Do not register debugfs as filesystem"
587 The API is open but filesystem is not loaded. Clients can still do
588 their work and read with debug tools that do not need
591 config DEBUG_FS_ALLOW_NONE
594 Access is off. Clients get -PERM when trying to create nodes in
595 debugfs tree and debugfs is not registered as a filesystem.
596 Client can then back-off or continue without debugfs access.
600 source "lib/Kconfig.kgdb"
601 source "lib/Kconfig.ubsan"
602 source "lib/Kconfig.kcsan"
607 bool "Kernel debugging"
609 Say Y here if you are developing drivers or trying to debug and
610 identify kernel problems.
613 bool "Miscellaneous debug code"
615 depends on DEBUG_KERNEL
617 Say Y here if you need to enable miscellaneous debug code that should
618 be under a more specific debug option but isn't.
620 menu "Networking Debugging"
622 source "net/Kconfig.debug"
624 endmenu # "Networking Debugging"
626 menu "Memory Debugging"
628 source "mm/Kconfig.debug"
631 bool "Debug object operations"
632 depends on DEBUG_KERNEL
634 If you say Y here, additional code will be inserted into the
635 kernel to track the life time of various objects and validate
636 the operations on those objects.
638 config DEBUG_OBJECTS_SELFTEST
639 bool "Debug objects selftest"
640 depends on DEBUG_OBJECTS
642 This enables the selftest of the object debug code.
644 config DEBUG_OBJECTS_FREE
645 bool "Debug objects in freed memory"
646 depends on DEBUG_OBJECTS
648 This enables checks whether a k/v free operation frees an area
649 which contains an object which has not been deactivated
650 properly. This can make kmalloc/kfree-intensive workloads
653 config DEBUG_OBJECTS_TIMERS
654 bool "Debug timer objects"
655 depends on DEBUG_OBJECTS
657 If you say Y here, additional code will be inserted into the
658 timer routines to track the life time of timer objects and
659 validate the timer operations.
661 config DEBUG_OBJECTS_WORK
662 bool "Debug work objects"
663 depends on DEBUG_OBJECTS
665 If you say Y here, additional code will be inserted into the
666 work queue routines to track the life time of work objects and
667 validate the work operations.
669 config DEBUG_OBJECTS_RCU_HEAD
670 bool "Debug RCU callbacks objects"
671 depends on DEBUG_OBJECTS
673 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
675 config DEBUG_OBJECTS_PERCPU_COUNTER
676 bool "Debug percpu counter objects"
677 depends on DEBUG_OBJECTS
679 If you say Y here, additional code will be inserted into the
680 percpu counter routines to track the life time of percpu counter
681 objects and validate the percpu counter operations.
683 config DEBUG_OBJECTS_ENABLE_DEFAULT
684 int "debug_objects bootup default value (0-1)"
687 depends on DEBUG_OBJECTS
689 Debug objects boot parameter default value
692 bool "Debug slab memory allocations"
693 depends on DEBUG_KERNEL && SLAB
695 Say Y here to have the kernel do limited verification on memory
696 allocation as well as poisoning memory on free to catch use of freed
697 memory. This can make kmalloc/kfree-intensive workloads much slower.
700 bool "SLUB debugging on by default"
701 depends on SLUB && SLUB_DEBUG
704 Boot with debugging on by default. SLUB boots by default with
705 the runtime debug capabilities switched off. Enabling this is
706 equivalent to specifying the "slub_debug" parameter on boot.
707 There is no support for more fine grained debug control like
708 possible with slub_debug=xxx. SLUB debugging may be switched
709 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
714 bool "Enable SLUB performance statistics"
715 depends on SLUB && SYSFS
717 SLUB statistics are useful to debug SLUBs allocation behavior in
718 order find ways to optimize the allocator. This should never be
719 enabled for production use since keeping statistics slows down
720 the allocator by a few percentage points. The slabinfo command
721 supports the determination of the most active slabs to figure
722 out which slabs are relevant to a particular load.
723 Try running: slabinfo -DA
725 config HAVE_DEBUG_KMEMLEAK
728 config DEBUG_KMEMLEAK
729 bool "Kernel memory leak detector"
730 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
732 select STACKTRACE if STACKTRACE_SUPPORT
736 Say Y here if you want to enable the memory leak
737 detector. The memory allocation/freeing is traced in a way
738 similar to the Boehm's conservative garbage collector, the
739 difference being that the orphan objects are not freed but
740 only shown in /sys/kernel/debug/kmemleak. Enabling this
741 feature will introduce an overhead to memory
742 allocations. See Documentation/dev-tools/kmemleak.rst for more
745 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
746 of finding leaks due to the slab objects poisoning.
748 In order to access the kmemleak file, debugfs needs to be
749 mounted (usually at /sys/kernel/debug).
751 config DEBUG_KMEMLEAK_MEM_POOL_SIZE
752 int "Kmemleak memory pool size"
753 depends on DEBUG_KMEMLEAK
757 Kmemleak must track all the memory allocations to avoid
758 reporting false positives. Since memory may be allocated or
759 freed before kmemleak is fully initialised, use a static pool
760 of metadata objects to track such callbacks. After kmemleak is
761 fully initialised, this memory pool acts as an emergency one
762 if slab allocations fail.
764 config DEBUG_KMEMLEAK_TEST
765 tristate "Simple test for the kernel memory leak detector"
766 depends on DEBUG_KMEMLEAK && m
768 This option enables a module that explicitly leaks memory.
772 config DEBUG_KMEMLEAK_DEFAULT_OFF
773 bool "Default kmemleak to off"
774 depends on DEBUG_KMEMLEAK
776 Say Y here to disable kmemleak by default. It can then be enabled
777 on the command line via kmemleak=on.
779 config DEBUG_KMEMLEAK_AUTO_SCAN
780 bool "Enable kmemleak auto scan thread on boot up"
782 depends on DEBUG_KMEMLEAK
784 Depending on the cpu, kmemleak scan may be cpu intensive and can
785 stall user tasks at times. This option enables/disables automatic
786 kmemleak scan at boot up.
788 Say N here to disable kmemleak auto scan thread to stop automatic
789 scanning. Disabling this option disables automatic reporting of
794 config DEBUG_STACK_USAGE
795 bool "Stack utilization instrumentation"
796 depends on DEBUG_KERNEL && !IA64
798 Enables the display of the minimum amount of free stack which each
799 task has ever had available in the sysrq-T and sysrq-P debug output.
801 This option will slow down process creation somewhat.
803 config SCHED_STACK_END_CHECK
804 bool "Detect stack corruption on calls to schedule()"
805 depends on DEBUG_KERNEL
808 This option checks for a stack overrun on calls to schedule().
809 If the stack end location is found to be over written always panic as
810 the content of the corrupted region can no longer be trusted.
811 This is to ensure no erroneous behaviour occurs which could result in
812 data corruption or a sporadic crash at a later stage once the region
813 is examined. The runtime overhead introduced is minimal.
815 config ARCH_HAS_DEBUG_VM_PGTABLE
818 An architecture should select this when it can successfully
819 build and run DEBUG_VM_PGTABLE.
823 depends on DEBUG_KERNEL
825 Enable this to turn on extended checks in the virtual-memory system
826 that may impact performance.
830 config DEBUG_VM_VMACACHE
831 bool "Debug VMA caching"
834 Enable this to turn on VMA caching debug information. Doing so
835 can cause significant overhead, so only enable it in non-production
841 bool "Debug VM red-black trees"
844 Enable VM red-black tree debugging information and extra validations.
848 config DEBUG_VM_PGFLAGS
849 bool "Debug page-flags operations"
852 Enables extra validation on page flags operations.
856 config DEBUG_VM_PGTABLE
857 bool "Debug arch page table for semantics compliance"
859 depends on ARCH_HAS_DEBUG_VM_PGTABLE
860 default y if DEBUG_VM
862 This option provides a debug method which can be used to test
863 architecture page table helper functions on various platforms in
864 verifying if they comply with expected generic MM semantics. This
865 will help architecture code in making sure that any changes or
866 new additions of these helpers still conform to expected
867 semantics of the generic MM. Platforms will have to opt in for
868 this through ARCH_HAS_DEBUG_VM_PGTABLE.
872 config ARCH_HAS_DEBUG_VIRTUAL
876 bool "Debug VM translations"
877 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL
879 Enable some costly sanity checks in virtual to page code. This can
880 catch mistakes with virt_to_page() and friends.
884 config DEBUG_NOMMU_REGIONS
885 bool "Debug the global anon/private NOMMU mapping region tree"
886 depends on DEBUG_KERNEL && !MMU
888 This option causes the global tree of anonymous and private mapping
889 regions to be regularly checked for invalid topology.
891 config DEBUG_MEMORY_INIT
892 bool "Debug memory initialisation" if EXPERT
895 Enable this for additional checks during memory initialisation.
896 The sanity checks verify aspects of the VM such as the memory model
897 and other information provided by the architecture. Verbose
898 information will be printed at KERN_DEBUG loglevel depending
899 on the mminit_loglevel= command-line option.
903 config MEMORY_NOTIFIER_ERROR_INJECT
904 tristate "Memory hotplug notifier error injection module"
905 depends on MEMORY_HOTPLUG && NOTIFIER_ERROR_INJECTION
907 This option provides the ability to inject artificial errors to
908 memory hotplug notifier chain callbacks. It is controlled through
909 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
911 If the notifier call chain should be failed with some events
912 notified, write the error code to "actions/<notifier event>/error".
914 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
916 # cd /sys/kernel/debug/notifier-error-inject/memory
917 # echo -12 > actions/MEM_GOING_OFFLINE/error
918 # echo offline > /sys/devices/system/memory/memoryXXX/state
919 bash: echo: write error: Cannot allocate memory
921 To compile this code as a module, choose M here: the module will
922 be called memory-notifier-error-inject.
926 config DEBUG_PER_CPU_MAPS
927 bool "Debug access to per_cpu maps"
928 depends on DEBUG_KERNEL
931 Say Y to verify that the per_cpu map being accessed has
932 been set up. This adds a fair amount of code to kernel memory
933 and decreases performance.
937 config DEBUG_KMAP_LOCAL
938 bool "Debug kmap_local temporary mappings"
939 depends on DEBUG_KERNEL && KMAP_LOCAL
941 This option enables additional error checking for the kmap_local
942 infrastructure. Disable for production use.
944 config ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
947 config DEBUG_KMAP_LOCAL_FORCE_MAP
948 bool "Enforce kmap_local temporary mappings"
949 depends on DEBUG_KERNEL && ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
951 select DEBUG_KMAP_LOCAL
953 This option enforces temporary mappings through the kmap_local
954 mechanism for non-highmem pages and on non-highmem systems.
955 Disable this for production systems!
958 bool "Highmem debugging"
959 depends on DEBUG_KERNEL && HIGHMEM
960 select DEBUG_KMAP_LOCAL_FORCE_MAP if ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
961 select DEBUG_KMAP_LOCAL
963 This option enables additional error checking for high memory
964 systems. Disable for production systems.
966 config HAVE_DEBUG_STACKOVERFLOW
969 config DEBUG_STACKOVERFLOW
970 bool "Check for stack overflows"
971 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
973 Say Y here if you want to check for overflows of kernel, IRQ
974 and exception stacks (if your architecture uses them). This
975 option will show detailed messages if free stack space drops
976 below a certain limit.
978 These kinds of bugs usually occur when call-chains in the
979 kernel get too deep, especially when interrupts are
982 Use this in cases where you see apparently random memory
983 corruption, especially if it appears in 'struct thread_info'
985 If in doubt, say "N".
987 source "lib/Kconfig.kasan"
988 source "lib/Kconfig.kfence"
990 endmenu # "Memory Debugging"
993 bool "Debug shared IRQ handlers"
994 depends on DEBUG_KERNEL
996 Enable this to generate a spurious interrupt just before a shared
997 interrupt handler is deregistered (generating one when registering
998 is currently disabled). Drivers need to handle this correctly. Some
999 don't and need to be caught.
1001 menu "Debug Oops, Lockups and Hangs"
1003 config PANIC_ON_OOPS
1004 bool "Panic on Oops"
1006 Say Y here to enable the kernel to panic when it oopses. This
1007 has the same effect as setting oops=panic on the kernel command
1010 This feature is useful to ensure that the kernel does not do
1011 anything erroneous after an oops which could result in data
1012 corruption or other issues.
1016 config PANIC_ON_OOPS_VALUE
1019 default 0 if !PANIC_ON_OOPS
1020 default 1 if PANIC_ON_OOPS
1022 config PANIC_TIMEOUT
1026 Set the timeout value (in seconds) until a reboot occurs when
1027 the kernel panics. If n = 0, then we wait forever. A timeout
1028 value n > 0 will wait n seconds before rebooting, while a timeout
1029 value n < 0 will reboot immediately.
1031 config LOCKUP_DETECTOR
1034 config SOFTLOCKUP_DETECTOR
1035 bool "Detect Soft Lockups"
1036 depends on DEBUG_KERNEL && !S390
1037 select LOCKUP_DETECTOR
1039 Say Y here to enable the kernel to act as a watchdog to detect
1042 Softlockups are bugs that cause the kernel to loop in kernel
1043 mode for more than 20 seconds, without giving other tasks a
1044 chance to run. The current stack trace is displayed upon
1045 detection and the system will stay locked up.
1047 config BOOTPARAM_SOFTLOCKUP_PANIC
1048 bool "Panic (Reboot) On Soft Lockups"
1049 depends on SOFTLOCKUP_DETECTOR
1051 Say Y here to enable the kernel to panic on "soft lockups",
1052 which are bugs that cause the kernel to loop in kernel
1053 mode for more than 20 seconds (configurable using the watchdog_thresh
1054 sysctl), without giving other tasks a chance to run.
1056 The panic can be used in combination with panic_timeout,
1057 to cause the system to reboot automatically after a
1058 lockup has been detected. This feature is useful for
1059 high-availability systems that have uptime guarantees and
1060 where a lockup must be resolved ASAP.
1064 config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
1066 depends on SOFTLOCKUP_DETECTOR
1068 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
1069 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
1071 config HARDLOCKUP_DETECTOR_PERF
1073 select SOFTLOCKUP_DETECTOR
1076 # Enables a timestamp based low pass filter to compensate for perf based
1077 # hard lockup detection which runs too fast due to turbo modes.
1079 config HARDLOCKUP_CHECK_TIMESTAMP
1083 # arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard
1084 # lockup detector rather than the perf based detector.
1086 config HARDLOCKUP_DETECTOR
1087 bool "Detect Hard Lockups"
1088 depends on DEBUG_KERNEL && !S390
1089 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH
1090 select LOCKUP_DETECTOR
1091 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF
1093 Say Y here to enable the kernel to act as a watchdog to detect
1096 Hardlockups are bugs that cause the CPU to loop in kernel mode
1097 for more than 10 seconds, without letting other interrupts have a
1098 chance to run. The current stack trace is displayed upon detection
1099 and the system will stay locked up.
1101 config BOOTPARAM_HARDLOCKUP_PANIC
1102 bool "Panic (Reboot) On Hard Lockups"
1103 depends on HARDLOCKUP_DETECTOR
1105 Say Y here to enable the kernel to panic on "hard lockups",
1106 which are bugs that cause the kernel to loop in kernel
1107 mode with interrupts disabled for more than 10 seconds (configurable
1108 using the watchdog_thresh sysctl).
1112 config BOOTPARAM_HARDLOCKUP_PANIC_VALUE
1114 depends on HARDLOCKUP_DETECTOR
1116 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC
1117 default 1 if BOOTPARAM_HARDLOCKUP_PANIC
1119 config DETECT_HUNG_TASK
1120 bool "Detect Hung Tasks"
1121 depends on DEBUG_KERNEL
1122 default SOFTLOCKUP_DETECTOR
1124 Say Y here to enable the kernel to detect "hung tasks",
1125 which are bugs that cause the task to be stuck in
1126 uninterruptible "D" state indefinitely.
1128 When a hung task is detected, the kernel will print the
1129 current stack trace (which you should report), but the
1130 task will stay in uninterruptible state. If lockdep is
1131 enabled then all held locks will also be reported. This
1132 feature has negligible overhead.
1134 config DEFAULT_HUNG_TASK_TIMEOUT
1135 int "Default timeout for hung task detection (in seconds)"
1136 depends on DETECT_HUNG_TASK
1139 This option controls the default timeout (in seconds) used
1140 to determine when a task has become non-responsive and should
1143 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
1144 sysctl or by writing a value to
1145 /proc/sys/kernel/hung_task_timeout_secs.
1147 A timeout of 0 disables the check. The default is two minutes.
1148 Keeping the default should be fine in most cases.
1150 config BOOTPARAM_HUNG_TASK_PANIC
1151 bool "Panic (Reboot) On Hung Tasks"
1152 depends on DETECT_HUNG_TASK
1154 Say Y here to enable the kernel to panic on "hung tasks",
1155 which are bugs that cause the kernel to leave a task stuck
1156 in uninterruptible "D" state.
1158 The panic can be used in combination with panic_timeout,
1159 to cause the system to reboot automatically after a
1160 hung task has been detected. This feature is useful for
1161 high-availability systems that have uptime guarantees and
1162 where a hung tasks must be resolved ASAP.
1166 config BOOTPARAM_HUNG_TASK_PANIC_VALUE
1168 depends on DETECT_HUNG_TASK
1170 default 0 if !BOOTPARAM_HUNG_TASK_PANIC
1171 default 1 if BOOTPARAM_HUNG_TASK_PANIC
1174 bool "Detect Workqueue Stalls"
1175 depends on DEBUG_KERNEL
1177 Say Y here to enable stall detection on workqueues. If a
1178 worker pool doesn't make forward progress on a pending work
1179 item for over a given amount of time, 30s by default, a
1180 warning message is printed along with dump of workqueue
1181 state. This can be configured through kernel parameter
1182 "workqueue.watchdog_thresh" and its sysfs counterpart.
1185 tristate "Test module to generate lockups"
1188 This builds the "test_lockup" module that helps to make sure
1189 that watchdogs and lockup detectors are working properly.
1191 Depending on module parameters it could emulate soft or hard
1192 lockup, "hung task", or locking arbitrary lock for a long time.
1193 Also it could generate series of lockups with cooling-down periods.
1197 endmenu # "Debug lockups and hangs"
1199 menu "Scheduler Debugging"
1202 bool "Collect scheduler debugging info"
1203 depends on DEBUG_KERNEL && PROC_FS
1206 If you say Y here, the /proc/sched_debug file will be provided
1207 that can help debug the scheduler. The runtime overhead of this
1215 bool "Collect scheduler statistics"
1216 depends on DEBUG_KERNEL && PROC_FS
1219 If you say Y here, additional code will be inserted into the
1220 scheduler and related routines to collect statistics about
1221 scheduler behavior and provide them in /proc/schedstat. These
1222 stats may be useful for both tuning and debugging the scheduler
1223 If you aren't debugging the scheduler or trying to tune a specific
1224 application, you can say N to avoid the very slight overhead
1229 config DEBUG_TIMEKEEPING
1230 bool "Enable extra timekeeping sanity checking"
1232 This option will enable additional timekeeping sanity checks
1233 which may be helpful when diagnosing issues where timekeeping
1234 problems are suspected.
1236 This may include checks in the timekeeping hotpaths, so this
1237 option may have a (very small) performance impact to some
1242 config DEBUG_PREEMPT
1243 bool "Debug preemptible kernel"
1244 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT
1247 If you say Y here then the kernel will use a debug variant of the
1248 commonly used smp_processor_id() function and will print warnings
1249 if kernel code uses it in a preemption-unsafe way. Also, the kernel
1250 will detect preemption count underflows.
1252 menu "Lock Debugging (spinlocks, mutexes, etc...)"
1254 config LOCK_DEBUGGING_SUPPORT
1256 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1259 config PROVE_LOCKING
1260 bool "Lock debugging: prove locking correctness"
1261 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1263 select DEBUG_SPINLOCK
1264 select DEBUG_MUTEXES if !PREEMPT_RT
1265 select DEBUG_RT_MUTEXES if RT_MUTEXES
1267 select DEBUG_WW_MUTEX_SLOWPATH
1268 select DEBUG_LOCK_ALLOC
1269 select PREEMPT_COUNT if !ARCH_NO_PREEMPT
1270 select TRACE_IRQFLAGS
1273 This feature enables the kernel to prove that all locking
1274 that occurs in the kernel runtime is mathematically
1275 correct: that under no circumstance could an arbitrary (and
1276 not yet triggered) combination of observed locking
1277 sequences (on an arbitrary number of CPUs, running an
1278 arbitrary number of tasks and interrupt contexts) cause a
1281 In short, this feature enables the kernel to report locking
1282 related deadlocks before they actually occur.
1284 The proof does not depend on how hard and complex a
1285 deadlock scenario would be to trigger: how many
1286 participant CPUs, tasks and irq-contexts would be needed
1287 for it to trigger. The proof also does not depend on
1288 timing: if a race and a resulting deadlock is possible
1289 theoretically (no matter how unlikely the race scenario
1290 is), it will be proven so and will immediately be
1291 reported by the kernel (once the event is observed that
1292 makes the deadlock theoretically possible).
1294 If a deadlock is impossible (i.e. the locking rules, as
1295 observed by the kernel, are mathematically correct), the
1296 kernel reports nothing.
1298 NOTE: this feature can also be enabled for rwlocks, mutexes
1299 and rwsems - in which case all dependencies between these
1300 different locking variants are observed and mapped too, and
1301 the proof of observed correctness is also maintained for an
1302 arbitrary combination of these separate locking variants.
1304 For more details, see Documentation/locking/lockdep-design.rst.
1306 config PROVE_RAW_LOCK_NESTING
1307 bool "Enable raw_spinlock - spinlock nesting checks"
1308 depends on PROVE_LOCKING
1311 Enable the raw_spinlock vs. spinlock nesting checks which ensure
1312 that the lock nesting rules for PREEMPT_RT enabled kernels are
1315 NOTE: There are known nesting problems. So if you enable this
1316 option expect lockdep splats until these problems have been fully
1317 addressed which is work in progress. This config switch allows to
1318 identify and analyze these problems. It will be removed and the
1319 check permanently enabled once the main issues have been fixed.
1321 If unsure, select N.
1324 bool "Lock usage statistics"
1325 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1327 select DEBUG_SPINLOCK
1328 select DEBUG_MUTEXES if !PREEMPT_RT
1329 select DEBUG_RT_MUTEXES if RT_MUTEXES
1330 select DEBUG_LOCK_ALLOC
1333 This feature enables tracking lock contention points
1335 For more details, see Documentation/locking/lockstat.rst
1337 This also enables lock events required by "perf lock",
1339 If you want to use "perf lock", you also need to turn on
1340 CONFIG_EVENT_TRACING.
1342 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1343 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1345 config DEBUG_RT_MUTEXES
1346 bool "RT Mutex debugging, deadlock detection"
1347 depends on DEBUG_KERNEL && RT_MUTEXES
1349 This allows rt mutex semantics violations and rt mutex related
1350 deadlocks (lockups) to be detected and reported automatically.
1352 config DEBUG_SPINLOCK
1353 bool "Spinlock and rw-lock debugging: basic checks"
1354 depends on DEBUG_KERNEL
1355 select UNINLINE_SPIN_UNLOCK
1357 Say Y here and build SMP to catch missing spinlock initialization
1358 and certain other kinds of spinlock errors commonly made. This is
1359 best used in conjunction with the NMI watchdog so that spinlock
1360 deadlocks are also debuggable.
1362 config DEBUG_MUTEXES
1363 bool "Mutex debugging: basic checks"
1364 depends on DEBUG_KERNEL && !PREEMPT_RT
1366 This feature allows mutex semantics violations to be detected and
1369 config DEBUG_WW_MUTEX_SLOWPATH
1370 bool "Wait/wound mutex debugging: Slowpath testing"
1371 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1372 select DEBUG_LOCK_ALLOC
1373 select DEBUG_SPINLOCK
1374 select DEBUG_MUTEXES if !PREEMPT_RT
1375 select DEBUG_RT_MUTEXES if PREEMPT_RT
1377 This feature enables slowpath testing for w/w mutex users by
1378 injecting additional -EDEADLK wound/backoff cases. Together with
1379 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1380 will test all possible w/w mutex interface abuse with the
1381 exception of simply not acquiring all the required locks.
1382 Note that this feature can introduce significant overhead, so
1383 it really should not be enabled in a production or distro kernel,
1384 even a debug kernel. If you are a driver writer, enable it. If
1385 you are a distro, do not.
1388 bool "RW Semaphore debugging: basic checks"
1389 depends on DEBUG_KERNEL
1391 This debugging feature allows mismatched rw semaphore locks
1392 and unlocks to be detected and reported.
1394 config DEBUG_LOCK_ALLOC
1395 bool "Lock debugging: detect incorrect freeing of live locks"
1396 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1397 select DEBUG_SPINLOCK
1398 select DEBUG_MUTEXES if !PREEMPT_RT
1399 select DEBUG_RT_MUTEXES if RT_MUTEXES
1402 This feature will check whether any held lock (spinlock, rwlock,
1403 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1404 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1405 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1406 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1407 held during task exit.
1411 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1416 config LOCKDEP_SMALL
1420 int "Bitsize for MAX_LOCKDEP_ENTRIES"
1421 depends on LOCKDEP && !LOCKDEP_SMALL
1425 Try increasing this value if you hit "BUG: MAX_LOCKDEP_ENTRIES too low!" message.
1427 config LOCKDEP_CHAINS_BITS
1428 int "Bitsize for MAX_LOCKDEP_CHAINS"
1429 depends on LOCKDEP && !LOCKDEP_SMALL
1433 Try increasing this value if you hit "BUG: MAX_LOCKDEP_CHAINS too low!" message.
1435 config LOCKDEP_STACK_TRACE_BITS
1436 int "Bitsize for MAX_STACK_TRACE_ENTRIES"
1437 depends on LOCKDEP && !LOCKDEP_SMALL
1441 Try increasing this value if you hit "BUG: MAX_STACK_TRACE_ENTRIES too low!" message.
1443 config LOCKDEP_STACK_TRACE_HASH_BITS
1444 int "Bitsize for STACK_TRACE_HASH_SIZE"
1445 depends on LOCKDEP && !LOCKDEP_SMALL
1449 Try increasing this value if you need large MAX_STACK_TRACE_ENTRIES.
1451 config LOCKDEP_CIRCULAR_QUEUE_BITS
1452 int "Bitsize for elements in circular_queue struct"
1457 Try increasing this value if you hit "lockdep bfs error:-1" warning due to __cq_enqueue() failure.
1459 config DEBUG_LOCKDEP
1460 bool "Lock dependency engine debugging"
1461 depends on DEBUG_KERNEL && LOCKDEP
1462 select DEBUG_IRQFLAGS
1464 If you say Y here, the lock dependency engine will do
1465 additional runtime checks to debug itself, at the price
1466 of more runtime overhead.
1468 config DEBUG_ATOMIC_SLEEP
1469 bool "Sleep inside atomic section checking"
1470 select PREEMPT_COUNT
1471 depends on DEBUG_KERNEL
1472 depends on !ARCH_NO_PREEMPT
1474 If you say Y here, various routines which may sleep will become very
1475 noisy if they are called inside atomic sections: when a spinlock is
1476 held, inside an rcu read side critical section, inside preempt disabled
1477 sections, inside an interrupt, etc...
1479 config DEBUG_LOCKING_API_SELFTESTS
1480 bool "Locking API boot-time self-tests"
1481 depends on DEBUG_KERNEL
1483 Say Y here if you want the kernel to run a short self-test during
1484 bootup. The self-test checks whether common types of locking bugs
1485 are detected by debugging mechanisms or not. (if you disable
1486 lock debugging then those bugs won't be detected of course.)
1487 The following locking APIs are covered: spinlocks, rwlocks,
1490 config LOCK_TORTURE_TEST
1491 tristate "torture tests for locking"
1492 depends on DEBUG_KERNEL
1495 This option provides a kernel module that runs torture tests
1496 on kernel locking primitives. The kernel module may be built
1497 after the fact on the running kernel to be tested, if desired.
1499 Say Y here if you want kernel locking-primitive torture tests
1500 to be built into the kernel.
1501 Say M if you want these torture tests to build as a module.
1502 Say N if you are unsure.
1504 config WW_MUTEX_SELFTEST
1505 tristate "Wait/wound mutex selftests"
1507 This option provides a kernel module that runs tests on the
1508 on the struct ww_mutex locking API.
1510 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction
1511 with this test harness.
1513 Say M if you want these self tests to build as a module.
1514 Say N if you are unsure.
1516 config SCF_TORTURE_TEST
1517 tristate "torture tests for smp_call_function*()"
1518 depends on DEBUG_KERNEL
1521 This option provides a kernel module that runs torture tests
1522 on the smp_call_function() family of primitives. The kernel
1523 module may be built after the fact on the running kernel to
1524 be tested, if desired.
1526 config CSD_LOCK_WAIT_DEBUG
1527 bool "Debugging for csd_lock_wait(), called from smp_call_function*()"
1528 depends on DEBUG_KERNEL
1532 This option enables debug prints when CPUs are slow to respond
1533 to the smp_call_function*() IPI wrappers. These debug prints
1534 include the IPI handler function currently executing (if any)
1535 and relevant stack traces.
1537 endmenu # lock debugging
1539 config TRACE_IRQFLAGS
1540 depends on TRACE_IRQFLAGS_SUPPORT
1543 Enables hooks to interrupt enabling and disabling for
1544 either tracing or lock debugging.
1546 config TRACE_IRQFLAGS_NMI
1548 depends on TRACE_IRQFLAGS
1549 depends on TRACE_IRQFLAGS_NMI_SUPPORT
1551 config DEBUG_IRQFLAGS
1552 bool "Debug IRQ flag manipulation"
1554 Enables checks for potentially unsafe enabling or disabling of
1555 interrupts, such as calling raw_local_irq_restore() when interrupts
1559 bool "Stack backtrace support"
1560 depends on STACKTRACE_SUPPORT
1562 This option causes the kernel to create a /proc/pid/stack for
1563 every process, showing its current stack trace.
1564 It is also used by various kernel debugging features that require
1565 stack trace generation.
1567 config WARN_ALL_UNSEEDED_RANDOM
1568 bool "Warn for all uses of unseeded randomness"
1571 Some parts of the kernel contain bugs relating to their use of
1572 cryptographically secure random numbers before it's actually possible
1573 to generate those numbers securely. This setting ensures that these
1574 flaws don't go unnoticed, by enabling a message, should this ever
1575 occur. This will allow people with obscure setups to know when things
1576 are going wrong, so that they might contact developers about fixing
1579 Unfortunately, on some models of some architectures getting
1580 a fully seeded CRNG is extremely difficult, and so this can
1581 result in dmesg getting spammed for a surprisingly long
1582 time. This is really bad from a security perspective, and
1583 so architecture maintainers really need to do what they can
1584 to get the CRNG seeded sooner after the system is booted.
1585 However, since users cannot do anything actionable to
1586 address this, by default the kernel will issue only a single
1587 warning for the first use of unseeded randomness.
1589 Say Y here if you want to receive warnings for all uses of
1590 unseeded randomness. This will be of use primarily for
1591 those developers interested in improving the security of
1592 Linux kernels running on their architecture (or
1595 config DEBUG_KOBJECT
1596 bool "kobject debugging"
1597 depends on DEBUG_KERNEL
1599 If you say Y here, some extra kobject debugging messages will be sent
1602 config DEBUG_KOBJECT_RELEASE
1603 bool "kobject release debugging"
1604 depends on DEBUG_OBJECTS_TIMERS
1606 kobjects are reference counted objects. This means that their
1607 last reference count put is not predictable, and the kobject can
1608 live on past the point at which a driver decides to drop it's
1609 initial reference to the kobject gained on allocation. An
1610 example of this would be a struct device which has just been
1613 However, some buggy drivers assume that after such an operation,
1614 the memory backing the kobject can be immediately freed. This
1615 goes completely against the principles of a refcounted object.
1617 If you say Y here, the kernel will delay the release of kobjects
1618 on the last reference count to improve the visibility of this
1619 kind of kobject release bug.
1621 config HAVE_DEBUG_BUGVERBOSE
1624 menu "Debug kernel data structures"
1627 bool "Debug linked list manipulation"
1628 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION
1630 Enable this to turn on extended checks in the linked-list
1636 bool "Debug priority linked list manipulation"
1637 depends on DEBUG_KERNEL
1639 Enable this to turn on extended checks in the priority-ordered
1640 linked-list (plist) walking routines. This checks the entire
1641 list multiple times during each manipulation.
1646 bool "Debug SG table operations"
1647 depends on DEBUG_KERNEL
1649 Enable this to turn on checks on scatter-gather tables. This can
1650 help find problems with drivers that do not properly initialize
1655 config DEBUG_NOTIFIERS
1656 bool "Debug notifier call chains"
1657 depends on DEBUG_KERNEL
1659 Enable this to turn on sanity checking for notifier call chains.
1660 This is most useful for kernel developers to make sure that
1661 modules properly unregister themselves from notifier chains.
1662 This is a relatively cheap check but if you care about maximum
1665 config BUG_ON_DATA_CORRUPTION
1666 bool "Trigger a BUG when data corruption is detected"
1669 Select this option if the kernel should BUG when it encounters
1670 data corruption in kernel memory structures when they get checked
1677 config DEBUG_CREDENTIALS
1678 bool "Debug credential management"
1679 depends on DEBUG_KERNEL
1681 Enable this to turn on some debug checking for credential
1682 management. The additional code keeps track of the number of
1683 pointers from task_structs to any given cred struct, and checks to
1684 see that this number never exceeds the usage count of the cred
1687 Furthermore, if SELinux is enabled, this also checks that the
1688 security pointer in the cred struct is never seen to be invalid.
1692 source "kernel/rcu/Kconfig.debug"
1694 config DEBUG_WQ_FORCE_RR_CPU
1695 bool "Force round-robin CPU selection for unbound work items"
1696 depends on DEBUG_KERNEL
1699 Workqueue used to implicitly guarantee that work items queued
1700 without explicit CPU specified are put on the local CPU. This
1701 guarantee is no longer true and while local CPU is still
1702 preferred work items may be put on foreign CPUs. Kernel
1703 parameter "workqueue.debug_force_rr_cpu" is added to force
1704 round-robin CPU selection to flush out usages which depend on the
1705 now broken guarantee. This config option enables the debug
1706 feature by default. When enabled, memory and cache locality will
1709 config CPU_HOTPLUG_STATE_CONTROL
1710 bool "Enable CPU hotplug state control"
1711 depends on DEBUG_KERNEL
1712 depends on HOTPLUG_CPU
1715 Allows to write steps between "offline" and "online" to the CPUs
1716 sysfs target file so states can be stepped granular. This is a debug
1717 option for now as the hotplug machinery cannot be stopped and
1718 restarted at arbitrary points yet.
1720 Say N if your are unsure.
1723 bool "Latency measuring infrastructure"
1724 depends on DEBUG_KERNEL
1725 depends on STACKTRACE_SUPPORT
1727 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
1733 Enable this option if you want to use the LatencyTOP tool
1734 to find out which userspace is blocking on what kernel operations.
1736 source "kernel/trace/Kconfig"
1738 config PROVIDE_OHCI1394_DMA_INIT
1739 bool "Remote debugging over FireWire early on boot"
1740 depends on PCI && X86
1742 If you want to debug problems which hang or crash the kernel early
1743 on boot and the crashing machine has a FireWire port, you can use
1744 this feature to remotely access the memory of the crashed machine
1745 over FireWire. This employs remote DMA as part of the OHCI1394
1746 specification which is now the standard for FireWire controllers.
1748 With remote DMA, you can monitor the printk buffer remotely using
1749 firescope and access all memory below 4GB using fireproxy from gdb.
1750 Even controlling a kernel debugger is possible using remote DMA.
1754 If ohci1394_dma=early is used as boot parameter, it will initialize
1755 all OHCI1394 controllers which are found in the PCI config space.
1757 As all changes to the FireWire bus such as enabling and disabling
1758 devices cause a bus reset and thereby disable remote DMA for all
1759 devices, be sure to have the cable plugged and FireWire enabled on
1760 the debugging host before booting the debug target for debugging.
1762 This code (~1k) is freed after boot. By then, the firewire stack
1763 in charge of the OHCI-1394 controllers should be used instead.
1765 See Documentation/core-api/debugging-via-ohci1394.rst for more information.
1767 source "samples/Kconfig"
1769 config ARCH_HAS_DEVMEM_IS_ALLOWED
1772 config STRICT_DEVMEM
1773 bool "Filter access to /dev/mem"
1774 depends on MMU && DEVMEM
1775 depends on ARCH_HAS_DEVMEM_IS_ALLOWED || GENERIC_LIB_DEVMEM_IS_ALLOWED
1776 default y if PPC || X86 || ARM64
1778 If this option is disabled, you allow userspace (root) access to all
1779 of memory, including kernel and userspace memory. Accidental
1780 access to this is obviously disastrous, but specific access can
1781 be used by people debugging the kernel. Note that with PAT support
1782 enabled, even in this case there are restrictions on /dev/mem
1783 use due to the cache aliasing requirements.
1785 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
1786 file only allows userspace access to PCI space and the BIOS code and
1787 data regions. This is sufficient for dosemu and X and all common
1792 config IO_STRICT_DEVMEM
1793 bool "Filter I/O access to /dev/mem"
1794 depends on STRICT_DEVMEM
1796 If this option is disabled, you allow userspace (root) access to all
1797 io-memory regardless of whether a driver is actively using that
1798 range. Accidental access to this is obviously disastrous, but
1799 specific access can be used by people debugging kernel drivers.
1801 If this option is switched on, the /dev/mem file only allows
1802 userspace access to *idle* io-memory ranges (see /proc/iomem) This
1803 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
1804 if the driver using a given range cannot be disabled.
1808 menu "$(SRCARCH) Debugging"
1810 source "arch/$(SRCARCH)/Kconfig.debug"
1814 menu "Kernel Testing and Coverage"
1816 source "lib/kunit/Kconfig"
1818 config NOTIFIER_ERROR_INJECTION
1819 tristate "Notifier error injection"
1820 depends on DEBUG_KERNEL
1823 This option provides the ability to inject artificial errors to
1824 specified notifier chain callbacks. It is useful to test the error
1825 handling of notifier call chain failures.
1829 config PM_NOTIFIER_ERROR_INJECT
1830 tristate "PM notifier error injection module"
1831 depends on PM && NOTIFIER_ERROR_INJECTION
1832 default m if PM_DEBUG
1834 This option provides the ability to inject artificial errors to
1835 PM notifier chain callbacks. It is controlled through debugfs
1836 interface /sys/kernel/debug/notifier-error-inject/pm
1838 If the notifier call chain should be failed with some events
1839 notified, write the error code to "actions/<notifier event>/error".
1841 Example: Inject PM suspend error (-12 = -ENOMEM)
1843 # cd /sys/kernel/debug/notifier-error-inject/pm/
1844 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1845 # echo mem > /sys/power/state
1846 bash: echo: write error: Cannot allocate memory
1848 To compile this code as a module, choose M here: the module will
1849 be called pm-notifier-error-inject.
1853 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1854 tristate "OF reconfig notifier error injection module"
1855 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1857 This option provides the ability to inject artificial errors to
1858 OF reconfig notifier chain callbacks. It is controlled
1859 through debugfs interface under
1860 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1862 If the notifier call chain should be failed with some events
1863 notified, write the error code to "actions/<notifier event>/error".
1865 To compile this code as a module, choose M here: the module will
1866 be called of-reconfig-notifier-error-inject.
1870 config NETDEV_NOTIFIER_ERROR_INJECT
1871 tristate "Netdev notifier error injection module"
1872 depends on NET && NOTIFIER_ERROR_INJECTION
1874 This option provides the ability to inject artificial errors to
1875 netdevice notifier chain callbacks. It is controlled through debugfs
1876 interface /sys/kernel/debug/notifier-error-inject/netdev
1878 If the notifier call chain should be failed with some events
1879 notified, write the error code to "actions/<notifier event>/error".
1881 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1883 # cd /sys/kernel/debug/notifier-error-inject/netdev
1884 # echo -22 > actions/NETDEV_CHANGEMTU/error
1885 # ip link set eth0 mtu 1024
1886 RTNETLINK answers: Invalid argument
1888 To compile this code as a module, choose M here: the module will
1889 be called netdev-notifier-error-inject.
1893 config FUNCTION_ERROR_INJECTION
1895 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES
1897 config FAULT_INJECTION
1898 bool "Fault-injection framework"
1899 depends on DEBUG_KERNEL
1901 Provide fault-injection framework.
1902 For more details, see Documentation/fault-injection/.
1905 bool "Fault-injection capability for kmalloc"
1906 depends on FAULT_INJECTION
1907 depends on SLAB || SLUB
1909 Provide fault-injection capability for kmalloc.
1911 config FAIL_PAGE_ALLOC
1912 bool "Fault-injection capability for alloc_pages()"
1913 depends on FAULT_INJECTION
1915 Provide fault-injection capability for alloc_pages().
1917 config FAULT_INJECTION_USERCOPY
1918 bool "Fault injection capability for usercopy functions"
1919 depends on FAULT_INJECTION
1921 Provides fault-injection capability to inject failures
1922 in usercopy functions (copy_from_user(), get_user(), ...).
1924 config FAIL_MAKE_REQUEST
1925 bool "Fault-injection capability for disk IO"
1926 depends on FAULT_INJECTION && BLOCK
1928 Provide fault-injection capability for disk IO.
1930 config FAIL_IO_TIMEOUT
1931 bool "Fault-injection capability for faking disk interrupts"
1932 depends on FAULT_INJECTION && BLOCK
1934 Provide fault-injection capability on end IO handling. This
1935 will make the block layer "forget" an interrupt as configured,
1936 thus exercising the error handling.
1938 Only works with drivers that use the generic timeout handling,
1939 for others it won't do anything.
1942 bool "Fault-injection capability for futexes"
1944 depends on FAULT_INJECTION && FUTEX
1946 Provide fault-injection capability for futexes.
1948 config FAULT_INJECTION_DEBUG_FS
1949 bool "Debugfs entries for fault-injection capabilities"
1950 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1952 Enable configuration of fault-injection capabilities via debugfs.
1954 config FAIL_FUNCTION
1955 bool "Fault-injection capability for functions"
1956 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION
1958 Provide function-based fault-injection capability.
1959 This will allow you to override a specific function with a return
1960 with given return value. As a result, function caller will see
1961 an error value and have to handle it. This is useful to test the
1962 error handling in various subsystems.
1964 config FAIL_MMC_REQUEST
1965 bool "Fault-injection capability for MMC IO"
1966 depends on FAULT_INJECTION_DEBUG_FS && MMC
1968 Provide fault-injection capability for MMC IO.
1969 This will make the mmc core return data errors. This is
1970 useful to test the error handling in the mmc block device
1971 and to test how the mmc host driver handles retries from
1975 bool "Fault-injection capability for SunRPC"
1976 depends on FAULT_INJECTION_DEBUG_FS && SUNRPC_DEBUG
1978 Provide fault-injection capability for SunRPC and
1981 config FAULT_INJECTION_STACKTRACE_FILTER
1982 bool "stacktrace filter for fault-injection capabilities"
1983 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1986 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
1988 Provide stacktrace filter for fault-injection capabilities
1990 config ARCH_HAS_KCOV
1993 An architecture should select this when it can successfully
1994 build and run with CONFIG_KCOV. This typically requires
1995 disabling instrumentation for some early boot code.
1997 config CC_HAS_SANCOV_TRACE_PC
1998 def_bool $(cc-option,-fsanitize-coverage=trace-pc)
2002 bool "Code coverage for fuzzing"
2003 depends on ARCH_HAS_KCOV
2004 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS
2005 depends on !ARCH_WANTS_NO_INSTR || STACK_VALIDATION || \
2006 GCC_VERSION >= 120000 || CLANG_VERSION >= 130000
2008 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC
2010 KCOV exposes kernel code coverage information in a form suitable
2011 for coverage-guided fuzzing (randomized testing).
2013 If RANDOMIZE_BASE is enabled, PC values will not be stable across
2014 different machines and across reboots. If you need stable PC values,
2015 disable RANDOMIZE_BASE.
2017 For more details, see Documentation/dev-tools/kcov.rst.
2019 config KCOV_ENABLE_COMPARISONS
2020 bool "Enable comparison operands collection by KCOV"
2022 depends on $(cc-option,-fsanitize-coverage=trace-cmp)
2024 KCOV also exposes operands of every comparison in the instrumented
2025 code along with operand sizes and PCs of the comparison instructions.
2026 These operands can be used by fuzzing engines to improve the quality
2027 of fuzzing coverage.
2029 config KCOV_INSTRUMENT_ALL
2030 bool "Instrument all code by default"
2034 If you are doing generic system call fuzzing (like e.g. syzkaller),
2035 then you will want to instrument the whole kernel and you should
2036 say y here. If you are doing more targeted fuzzing (like e.g.
2037 filesystem fuzzing with AFL) then you will want to enable coverage
2038 for more specific subsets of files, and should say n here.
2040 config KCOV_IRQ_AREA_SIZE
2041 hex "Size of interrupt coverage collection area in words"
2045 KCOV uses preallocated per-cpu areas to collect coverage from
2046 soft interrupts. This specifies the size of those areas in the
2047 number of unsigned long words.
2049 menuconfig RUNTIME_TESTING_MENU
2050 bool "Runtime Testing"
2053 if RUNTIME_TESTING_MENU
2056 tristate "Linux Kernel Dump Test Tool Module"
2059 This module enables testing of the different dumping mechanisms by
2060 inducing system failures at predefined crash points.
2061 If you don't need it: say N
2062 Choose M here to compile this code as a module. The module will be
2065 Documentation on how to use the module can be found in
2066 Documentation/fault-injection/provoke-crashes.rst
2068 config TEST_LIST_SORT
2069 tristate "Linked list sorting test" if !KUNIT_ALL_TESTS
2071 default KUNIT_ALL_TESTS
2073 Enable this to turn on 'list_sort()' function test. This test is
2074 executed only once during system boot (so affects only boot time),
2075 or at module load time.
2079 config TEST_MIN_HEAP
2080 tristate "Min heap test"
2081 depends on DEBUG_KERNEL || m
2083 Enable this to turn on min heap function tests. This test is
2084 executed only once during system boot (so affects only boot time),
2085 or at module load time.
2090 tristate "Array-based sort test" if !KUNIT_ALL_TESTS
2092 default KUNIT_ALL_TESTS
2094 This option enables the self-test function of 'sort()' at boot,
2095 or at module load time.
2100 tristate "64bit/32bit division and modulo test"
2101 depends on DEBUG_KERNEL || m
2103 Enable this to turn on 'do_div()' function test. This test is
2104 executed only once during system boot (so affects only boot time),
2105 or at module load time.
2109 config KPROBES_SANITY_TEST
2110 tristate "Kprobes sanity tests"
2111 depends on DEBUG_KERNEL
2115 This option provides for testing basic kprobes functionality on
2116 boot. Samples of kprobe and kretprobe are inserted and
2117 verified for functionality.
2119 Say N if you are unsure.
2121 config BACKTRACE_SELF_TEST
2122 tristate "Self test for the backtrace code"
2123 depends on DEBUG_KERNEL
2125 This option provides a kernel module that can be used to test
2126 the kernel stack backtrace code. This option is not useful
2127 for distributions or general kernels, but only for kernel
2128 developers working on architecture code.
2130 Note that if you want to also test saved backtraces, you will
2131 have to enable STACKTRACE as well.
2133 Say N if you are unsure.
2135 config TEST_REF_TRACKER
2136 tristate "Self test for reference tracker"
2137 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
2140 This option provides a kernel module performing tests
2141 using reference tracker infrastructure.
2143 Say N if you are unsure.
2146 tristate "Red-Black tree test"
2147 depends on DEBUG_KERNEL
2149 A benchmark measuring the performance of the rbtree library.
2150 Also includes rbtree invariant checks.
2152 config REED_SOLOMON_TEST
2153 tristate "Reed-Solomon library test"
2154 depends on DEBUG_KERNEL || m
2156 select REED_SOLOMON_ENC16
2157 select REED_SOLOMON_DEC16
2159 This option enables the self-test function of rslib at boot,
2160 or at module load time.
2164 config INTERVAL_TREE_TEST
2165 tristate "Interval tree test"
2166 depends on DEBUG_KERNEL
2167 select INTERVAL_TREE
2169 A benchmark measuring the performance of the interval tree library
2172 tristate "Per cpu operations test"
2173 depends on m && DEBUG_KERNEL
2175 Enable this option to build test module which validates per-cpu
2180 config ATOMIC64_SELFTEST
2181 tristate "Perform an atomic64_t self-test"
2183 Enable this option to test the atomic64_t functions at boot or
2184 at module load time.
2188 config ASYNC_RAID6_TEST
2189 tristate "Self test for hardware accelerated raid6 recovery"
2190 depends on ASYNC_RAID6_RECOV
2193 This is a one-shot self test that permutes through the
2194 recovery of all the possible two disk failure scenarios for a
2195 N-disk array. Recovery is performed with the asynchronous
2196 raid6 recovery routines, and will optionally use an offload
2197 engine if one is available.
2202 tristate "Test functions located in the hexdump module at runtime"
2204 config STRING_SELFTEST
2205 tristate "Test string functions at runtime"
2207 config TEST_STRING_HELPERS
2208 tristate "Test functions located in the string_helpers module at runtime"
2211 tristate "Test strscpy*() family of functions at runtime"
2214 tristate "Test kstrto*() family of functions at runtime"
2217 tristate "Test printf() family of functions at runtime"
2220 tristate "Test scanf() family of functions at runtime"
2223 tristate "Test bitmap_*() family of functions at runtime"
2225 Enable this option to test the bitmap functions at boot.
2230 tristate "Test functions located in the uuid module at runtime"
2233 tristate "Test the XArray code at runtime"
2235 config TEST_OVERFLOW
2236 tristate "Test check_*_overflow() functions at runtime"
2238 config TEST_RHASHTABLE
2239 tristate "Perform selftest on resizable hash table"
2241 Enable this option to test the rhashtable functions at boot.
2246 tristate "Perform selftest on siphash functions"
2248 Enable this option to test the kernel's siphash (<linux/siphash.h>) hash
2249 functions on boot (or module load).
2251 This is intended to help people writing architecture-specific
2252 optimized versions. If unsure, say N.
2255 tristate "Perform selftest on IDA functions"
2258 tristate "Perform selftest on priority array manager"
2261 Enable this option to test priority array manager on boot
2266 config TEST_IRQ_TIMINGS
2267 bool "IRQ timings selftest"
2268 depends on IRQ_TIMINGS
2270 Enable this option to test the irq timings code on boot.
2275 tristate "Test module loading with 'hello world' module"
2278 This builds the "test_module" module that emits "Hello, world"
2279 on printk when loaded. It is designed to be used for basic
2280 evaluation of the module loading subsystem (for example when
2281 validating module verification). It lacks any extra dependencies,
2282 and will not normally be loaded by the system unless explicitly
2288 tristate "Test module for compilation of bitops operations"
2291 This builds the "test_bitops" module that is much like the
2292 TEST_LKM module except that it does a basic exercise of the
2293 set/clear_bit macros and get_count_order/long to make sure there are
2294 no compiler warnings from C=1 sparse checker or -Wextra
2295 compilations. It has no dependencies and doesn't run or load unless
2296 explicitly requested by name. for example: modprobe test_bitops.
2301 tristate "Test module for stress/performance analysis of vmalloc allocator"
2306 This builds the "test_vmalloc" module that should be used for
2307 stress and performance analysis. So, any new change for vmalloc
2308 subsystem can be evaluated from performance and stability point
2313 config TEST_USER_COPY
2314 tristate "Test user/kernel boundary protections"
2317 This builds the "test_user_copy" module that runs sanity checks
2318 on the copy_to/from_user infrastructure, making sure basic
2319 user/kernel boundary testing is working. If it fails to load,
2320 a regression has been detected in the user/kernel memory boundary
2326 tristate "Test BPF filter functionality"
2329 This builds the "test_bpf" module that runs various test vectors
2330 against the BPF interpreter or BPF JIT compiler depending on the
2331 current setting. This is in particular useful for BPF JIT compiler
2332 development, but also to run regression tests against changes in
2333 the interpreter code. It also enables test stubs for eBPF maps and
2334 verifier used by user space verifier testsuite.
2338 config TEST_BLACKHOLE_DEV
2339 tristate "Test blackhole netdev functionality"
2342 This builds the "test_blackhole_dev" module that validates the
2343 data path through this blackhole netdev.
2347 config FIND_BIT_BENCHMARK
2348 tristate "Test find_bit functions"
2350 This builds the "test_find_bit" module that measure find_*_bit()
2351 functions performance.
2355 config TEST_FIRMWARE
2356 tristate "Test firmware loading via userspace interface"
2357 depends on FW_LOADER
2359 This builds the "test_firmware" module that creates a userspace
2360 interface for testing firmware loading. This can be used to
2361 control the triggering of firmware loading without needing an
2362 actual firmware-using device. The contents can be rechecked by
2368 tristate "sysctl test driver"
2369 depends on PROC_SYSCTL
2371 This builds the "test_sysctl" module. This driver enables to test the
2372 proc sysctl interfaces available to drivers safely without affecting
2373 production knobs which might alter system functionality.
2377 config BITFIELD_KUNIT
2378 tristate "KUnit test bitfield functions at runtime"
2381 Enable this option to test the bitfield functions at boot.
2383 KUnit tests run during boot and output the results to the debug log
2384 in TAP format (http://testanything.org/). Only useful for kernel devs
2385 running the KUnit test harness, and not intended for inclusion into a
2388 For more information on KUnit and unit tests in general please refer
2389 to the KUnit documentation in Documentation/dev-tools/kunit/.
2393 config HASH_KUNIT_TEST
2394 tristate "KUnit Test for integer hash functions" if !KUNIT_ALL_TESTS
2396 default KUNIT_ALL_TESTS
2398 Enable this option to test the kernel's string (<linux/stringhash.h>), and
2399 integer (<linux/hash.h>) hash functions on boot.
2401 KUnit tests run during boot and output the results to the debug log
2402 in TAP format (https://testanything.org/). Only useful for kernel devs
2403 running the KUnit test harness, and not intended for inclusion into a
2406 For more information on KUnit and unit tests in general please refer
2407 to the KUnit documentation in Documentation/dev-tools/kunit/.
2409 This is intended to help people writing architecture-specific
2410 optimized versions. If unsure, say N.
2412 config RESOURCE_KUNIT_TEST
2413 tristate "KUnit test for resource API"
2416 This builds the resource API unit test.
2417 Tests the logic of API provided by resource.c and ioport.h.
2418 For more information on KUnit and unit tests in general please refer
2419 to the KUnit documentation in Documentation/dev-tools/kunit/.
2423 config SYSCTL_KUNIT_TEST
2424 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS
2426 default KUNIT_ALL_TESTS
2428 This builds the proc sysctl unit test, which runs on boot.
2429 Tests the API contract and implementation correctness of sysctl.
2430 For more information on KUnit and unit tests in general please refer
2431 to the KUnit documentation in Documentation/dev-tools/kunit/.
2435 config LIST_KUNIT_TEST
2436 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS
2438 default KUNIT_ALL_TESTS
2440 This builds the linked list KUnit test suite.
2441 It tests that the API and basic functionality of the list_head type
2442 and associated macros.
2444 KUnit tests run during boot and output the results to the debug log
2445 in TAP format (https://testanything.org/). Only useful for kernel devs
2446 running the KUnit test harness, and not intended for inclusion into a
2449 For more information on KUnit and unit tests in general please refer
2450 to the KUnit documentation in Documentation/dev-tools/kunit/.
2454 config LINEAR_RANGES_TEST
2455 tristate "KUnit test for linear_ranges"
2457 select LINEAR_RANGES
2459 This builds the linear_ranges unit test, which runs on boot.
2460 Tests the linear_ranges logic correctness.
2461 For more information on KUnit and unit tests in general please refer
2462 to the KUnit documentation in Documentation/dev-tools/kunit/.
2466 config CMDLINE_KUNIT_TEST
2467 tristate "KUnit test for cmdline API"
2470 This builds the cmdline API unit test.
2471 Tests the logic of API provided by cmdline.c.
2472 For more information on KUnit and unit tests in general please refer
2473 to the KUnit documentation in Documentation/dev-tools/kunit/.
2478 tristate "KUnit test for bits.h"
2481 This builds the bits unit test.
2482 Tests the logic of macros defined in bits.h.
2483 For more information on KUnit and unit tests in general please refer
2484 to the KUnit documentation in Documentation/dev-tools/kunit/.
2488 config SLUB_KUNIT_TEST
2489 tristate "KUnit test for SLUB cache error detection" if !KUNIT_ALL_TESTS
2490 depends on SLUB_DEBUG && KUNIT
2491 default KUNIT_ALL_TESTS
2493 This builds SLUB allocator unit test.
2494 Tests SLUB cache debugging functionality.
2495 For more information on KUnit and unit tests in general please refer
2496 to the KUnit documentation in Documentation/dev-tools/kunit/.
2500 config RATIONAL_KUNIT_TEST
2501 tristate "KUnit test for rational.c" if !KUNIT_ALL_TESTS
2502 depends on KUNIT && RATIONAL
2503 default KUNIT_ALL_TESTS
2505 This builds the rational math unit test.
2506 For more information on KUnit and unit tests in general please refer
2507 to the KUnit documentation in Documentation/dev-tools/kunit/.
2511 config MEMCPY_KUNIT_TEST
2512 tristate "Test memcpy(), memmove(), and memset() functions at runtime" if !KUNIT_ALL_TESTS
2514 default KUNIT_ALL_TESTS
2516 Builds unit tests for memcpy(), memmove(), and memset() functions.
2517 For more information on KUnit and unit tests in general please refer
2518 to the KUnit documentation in Documentation/dev-tools/kunit/.
2523 tristate "udelay test driver"
2525 This builds the "udelay_test" module that helps to make sure
2526 that udelay() is working properly.
2530 config TEST_STATIC_KEYS
2531 tristate "Test static keys"
2534 Test the static key interfaces.
2539 tristate "kmod stress tester"
2541 depends on NETDEVICES && NET_CORE && INET # for TUN
2543 depends on PAGE_SIZE_LESS_THAN_256KB # for BTRFS
2549 Test the kernel's module loading mechanism: kmod. kmod implements
2550 support to load modules using the Linux kernel's usermode helper.
2551 This test provides a series of tests against kmod.
2553 Although technically you can either build test_kmod as a module or
2554 into the kernel we disallow building it into the kernel since
2555 it stress tests request_module() and this will very likely cause
2556 some issues by taking over precious threads available from other
2557 module load requests, ultimately this could be fatal.
2561 tools/testing/selftests/kmod/kmod.sh --help
2565 config TEST_DEBUG_VIRTUAL
2566 tristate "Test CONFIG_DEBUG_VIRTUAL feature"
2567 depends on DEBUG_VIRTUAL
2569 Test the kernel's ability to detect incorrect calls to
2570 virt_to_phys() done against the non-linear part of the
2571 kernel's virtual address map.
2575 config TEST_MEMCAT_P
2576 tristate "Test memcat_p() helper function"
2578 Test the memcat_p() helper for correctly merging two
2579 pointer arrays together.
2583 config TEST_LIVEPATCH
2584 tristate "Test livepatching"
2586 depends on DYNAMIC_DEBUG
2587 depends on LIVEPATCH
2590 Test kernel livepatching features for correctness. The tests will
2591 load test modules that will be livepatched in various scenarios.
2593 To run all the livepatching tests:
2595 make -C tools/testing/selftests TARGETS=livepatch run_tests
2597 Alternatively, individual tests may be invoked:
2599 tools/testing/selftests/livepatch/test-callbacks.sh
2600 tools/testing/selftests/livepatch/test-livepatch.sh
2601 tools/testing/selftests/livepatch/test-shadow-vars.sh
2606 tristate "Perform selftest on object aggreration manager"
2610 Enable this option to test object aggregation manager on boot
2614 config TEST_STACKINIT
2615 tristate "Test level of stack variable initialization"
2617 Test if the kernel is zero-initializing stack variables and
2618 padding. Coverage is controlled by compiler flags,
2619 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF,
2620 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL.
2625 tristate "Test heap/page initialization"
2627 Test if the kernel is zero-initializing heap and page allocations.
2628 This can be useful to test init_on_alloc and init_on_free features.
2633 tristate "Test HMM (Heterogeneous Memory Management)"
2634 depends on TRANSPARENT_HUGEPAGE
2635 depends on DEVICE_PRIVATE
2639 This is a pseudo device driver solely for testing HMM.
2640 Say M here if you want to build the HMM test module.
2641 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests.
2645 config TEST_FREE_PAGES
2646 tristate "Test freeing pages"
2648 Test that a memory leak does not occur due to a race between
2649 freeing a block of pages and a speculative page reference.
2650 Loading this module is safe if your kernel has the bug fixed.
2651 If the bug is not fixed, it will leak gigabytes of memory and
2652 probably OOM your system.
2655 tristate "Test floating point operations in kernel space"
2656 depends on X86 && !KCOV_INSTRUMENT_ALL
2658 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu
2659 which will trigger a sequence of floating point operations. This is used
2660 for self-testing floating point control register setting in
2665 config TEST_CLOCKSOURCE_WATCHDOG
2666 tristate "Test clocksource watchdog in kernel space"
2667 depends on CLOCKSOURCE_WATCHDOG
2669 Enable this option to create a kernel module that will trigger
2670 a test of the clocksource watchdog. This module may be loaded
2671 via modprobe or insmod in which case it will run upon being
2672 loaded, or it may be built in, in which case it will run
2677 endif # RUNTIME_TESTING_MENU
2679 config ARCH_USE_MEMTEST
2682 An architecture should select this when it uses early_memtest()
2683 during boot process.
2687 depends on ARCH_USE_MEMTEST
2689 This option adds a kernel parameter 'memtest', which allows memtest
2690 to be set and executed.
2691 memtest=0, mean disabled; -- default
2692 memtest=1, mean do 1 test pattern;
2694 memtest=17, mean do 17 test patterns.
2695 If you are unsure how to answer this question, answer N.
2699 config HYPERV_TESTING
2700 bool "Microsoft Hyper-V driver testing"
2702 depends on HYPERV && DEBUG_FS
2704 Select this option to enable Hyper-V vmbus testing.
2706 endmenu # "Kernel Testing and Coverage"
2708 source "Documentation/Kconfig"
2710 endmenu # Kernel hacking