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 # Clang is known to generate .{s,u}leb128 with symbol deltas with DWARF5, which
212 # some targets may not support: https://sourceware.org/bugzilla/show_bug.cgi?id=27215
213 config AS_HAS_NON_CONST_LEB128
214 def_bool $(as-instr,.uleb128 .Lexpr_end4 - .Lexpr_start3\n.Lexpr_start3:\n.Lexpr_end4:)
216 menu "Compile-time checks and compiler options"
219 bool "Compile the kernel with debug info"
220 depends on DEBUG_KERNEL && !COMPILE_TEST
222 If you say Y here the resulting kernel image will include
223 debugging info resulting in a larger kernel image.
224 This adds debug symbols to the kernel and modules (gcc -g), and
225 is needed if you intend to use kernel crashdump or binary object
226 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
227 Say Y here only if you plan to debug the kernel.
233 config DEBUG_INFO_REDUCED
234 bool "Reduce debugging information"
236 If you say Y here gcc is instructed to generate less debugging
237 information for structure types. This means that tools that
238 need full debugging information (like kgdb or systemtap) won't
239 be happy. But if you merely need debugging information to
240 resolve line numbers there is no loss. Advantage is that
241 build directory object sizes shrink dramatically over a full
242 DEBUG_INFO build and compile times are reduced too.
243 Only works with newer gcc versions.
245 config DEBUG_INFO_COMPRESSED
246 bool "Compressed debugging information"
247 depends on $(cc-option,-gz=zlib)
248 depends on $(ld-option,--compress-debug-sections=zlib)
250 Compress the debug information using zlib. Requires GCC 5.0+ or Clang
251 5.0+, binutils 2.26+, and zlib.
253 Users of dpkg-deb via scripts/package/builddeb may find an increase in
254 size of their debug .deb packages with this config set, due to the
255 debug info being compressed with zlib, then the object files being
256 recompressed with a different compression scheme. But this is still
257 preferable to setting $KDEB_COMPRESS to "none" which would be even
260 config DEBUG_INFO_SPLIT
261 bool "Produce split debuginfo in .dwo files"
262 depends on $(cc-option,-gsplit-dwarf)
264 Generate debug info into separate .dwo files. This significantly
265 reduces the build directory size for builds with DEBUG_INFO,
266 because it stores the information only once on disk in .dwo
267 files instead of multiple times in object files and executables.
268 In addition the debug information is also compressed.
270 Requires recent gcc (4.7+) and recent gdb/binutils.
271 Any tool that packages or reads debug information would need
272 to know about the .dwo files and include them.
273 Incompatible with older versions of ccache.
276 prompt "DWARF version"
278 Which version of DWARF debug info to emit.
280 config DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT
281 bool "Rely on the toolchain's implicit default DWARF version"
282 depends on !CC_IS_CLANG || AS_IS_LLVM || CLANG_VERSION < 140000 || (AS_IS_GNU && AS_VERSION >= 23502 && AS_HAS_NON_CONST_LEB128)
284 The implicit default version of DWARF debug info produced by a
285 toolchain changes over time.
287 This can break consumers of the debug info that haven't upgraded to
288 support newer revisions, and prevent testing newer versions, but
289 those should be less common scenarios.
293 config DEBUG_INFO_DWARF4
294 bool "Generate DWARF Version 4 debuginfo"
296 Generate DWARF v4 debug info. This requires gcc 4.5+ and gdb 7.0+.
298 If you have consumers of DWARF debug info that are not ready for
299 newer revisions of DWARF, you may wish to choose this or have your
302 config DEBUG_INFO_DWARF5
303 bool "Generate DWARF Version 5 debuginfo"
304 depends on !CC_IS_CLANG || AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502 && AS_HAS_NON_CONST_LEB128)
305 depends on !DEBUG_INFO_BTF || PAHOLE_VERSION >= 121
307 Generate DWARF v5 debug info. Requires binutils 2.35.2, gcc 5.0+ (gcc
308 5.0+ accepts the -gdwarf-5 flag but only had partial support for some
309 draft features until 7.0), and gdb 8.0+.
311 Changes to the structure of debug info in Version 5 allow for around
312 15-18% savings in resulting image and debug info section sizes as
313 compared to DWARF Version 4. DWARF Version 5 standardizes previous
314 extensions such as accelerators for symbol indexing and the format
315 for fission (.dwo/.dwp) files. Users may not want to select this
316 config if they rely on tooling that has not yet been updated to
317 support DWARF Version 5.
319 endchoice # "DWARF version"
321 config DEBUG_INFO_BTF
322 bool "Generate BTF typeinfo"
323 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED
324 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST
326 Generate deduplicated BTF type information from DWARF debug info.
327 Turning this on expects presence of pahole tool, which will convert
328 DWARF type info into equivalent deduplicated BTF type info.
330 config PAHOLE_HAS_SPLIT_BTF
331 def_bool PAHOLE_VERSION >= 119
333 config DEBUG_INFO_BTF_MODULES
335 depends on DEBUG_INFO_BTF && MODULES && PAHOLE_HAS_SPLIT_BTF
337 Generate compact split BTF type information for kernel modules.
340 bool "Provide GDB scripts for kernel debugging"
342 This creates the required links to GDB helper scripts in the
343 build directory. If you load vmlinux into gdb, the helper
344 scripts will be automatically imported by gdb as well, and
345 additional functions are available to analyze a Linux kernel
346 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst
352 int "Warn for stack frames larger than"
354 default 2048 if GCC_PLUGIN_LATENT_ENTROPY
355 default 2048 if PARISC
356 default 1536 if (!64BIT && XTENSA)
357 default 1280 if KASAN && !64BIT
358 default 1024 if !64BIT
359 default 2048 if 64BIT
361 Tell gcc to warn at build time for stack frames larger than this.
362 Setting this too low will cause a lot of warnings.
363 Setting it to 0 disables the warning.
365 config STRIP_ASM_SYMS
366 bool "Strip assembler-generated symbols during link"
369 Strip internal assembler-generated symbols during a link (symbols
370 that look like '.Lxxx') so they don't pollute the output of
371 get_wchan() and suchlike.
374 bool "Generate readable assembler code"
375 depends on DEBUG_KERNEL
378 Disable some compiler optimizations that tend to generate human unreadable
379 assembler output. This may make the kernel slightly slower, but it helps
380 to keep kernel developers who have to stare a lot at assembler listings
383 config HEADERS_INSTALL
384 bool "Install uapi headers to usr/include"
387 This option will install uapi headers (headers exported to user-space)
388 into the usr/include directory for use during the kernel build.
389 This is unneeded for building the kernel itself, but needed for some
390 user-space program samples. It is also needed by some features such
391 as uapi header sanity checks.
393 config DEBUG_SECTION_MISMATCH
394 bool "Enable full Section mismatch analysis"
397 The section mismatch analysis checks if there are illegal
398 references from one section to another section.
399 During linktime or runtime, some sections are dropped;
400 any use of code/data previously in these sections would
401 most likely result in an oops.
402 In the code, functions and variables are annotated with
403 __init,, etc. (see the full list in include/linux/init.h),
404 which results in the code/data being placed in specific sections.
405 The section mismatch analysis is always performed after a full
406 kernel build, and enabling this option causes the following
407 additional step to occur:
408 - Add the option -fno-inline-functions-called-once to gcc commands.
409 When inlining a function annotated with __init in a non-init
410 function, we would lose the section information and thus
411 the analysis would not catch the illegal reference.
412 This option tells gcc to inline less (but it does result in
415 config SECTION_MISMATCH_WARN_ONLY
416 bool "Make section mismatch errors non-fatal"
419 If you say N here, the build process will fail if there are any
420 section mismatch, instead of just throwing warnings.
424 config DEBUG_FORCE_FUNCTION_ALIGN_64B
425 bool "Force all function address 64B aligned"
426 depends on EXPERT && (X86_64 || ARM64 || PPC32 || PPC64 || ARC)
428 There are cases that a commit from one domain changes the function
429 address alignment of other domains, and cause magic performance
430 bump (regression or improvement). Enable this option will help to
431 verify if the bump is caused by function alignment changes, while
432 it will slightly increase the kernel size and affect icache usage.
434 It is mainly for debug and performance tuning use.
437 # Select this config option from the architecture Kconfig, if it
438 # is preferred to always offer frame pointers as a config
439 # option on the architecture (regardless of KERNEL_DEBUG):
441 config ARCH_WANT_FRAME_POINTERS
445 bool "Compile the kernel with frame pointers"
446 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS
447 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
449 If you say Y here the resulting kernel image will be slightly
450 larger and slower, but it gives very useful debugging information
451 in case of kernel bugs. (precise oopses/stacktraces/warnings)
453 config STACK_VALIDATION
454 bool "Compile-time stack metadata validation"
455 depends on HAVE_STACK_VALIDATION
458 Add compile-time checks to validate stack metadata, including frame
459 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure
460 that runtime stack traces are more reliable.
462 This is also a prerequisite for generation of ORC unwind data, which
463 is needed for CONFIG_UNWINDER_ORC.
465 For more information, see
466 tools/objtool/Documentation/stack-validation.txt.
468 config VMLINUX_VALIDATION
470 depends on STACK_VALIDATION && DEBUG_ENTRY && !PARAVIRT
474 bool "Generate vmlinux.map file when linking"
477 Selecting this option will pass "-Map=vmlinux.map" to ld
478 when linking vmlinux. That file can be useful for verifying
479 and debugging magic section games, and for seeing which
480 pieces of code get eliminated with
481 CONFIG_LD_DEAD_CODE_DATA_ELIMINATION.
483 config DEBUG_FORCE_WEAK_PER_CPU
484 bool "Force weak per-cpu definitions"
485 depends on DEBUG_KERNEL
487 s390 and alpha require percpu variables in modules to be
488 defined weak to work around addressing range issue which
489 puts the following two restrictions on percpu variable
492 1. percpu symbols must be unique whether static or not
493 2. percpu variables can't be defined inside a function
495 To ensure that generic code follows the above rules, this
496 option forces all percpu variables to be defined as weak.
498 endmenu # "Compiler options"
500 menu "Generic Kernel Debugging Instruments"
503 bool "Magic SysRq key"
506 If you say Y here, you will have some control over the system even
507 if the system crashes for example during kernel debugging (e.g., you
508 will be able to flush the buffer cache to disk, reboot the system
509 immediately or dump some status information). This is accomplished
510 by pressing various keys while holding SysRq (Alt+PrintScreen). It
511 also works on a serial console (on PC hardware at least), if you
512 send a BREAK and then within 5 seconds a command keypress. The
513 keys are documented in <file:Documentation/admin-guide/sysrq.rst>.
514 Don't say Y unless you really know what this hack does.
516 config MAGIC_SYSRQ_DEFAULT_ENABLE
517 hex "Enable magic SysRq key functions by default"
518 depends on MAGIC_SYSRQ
521 Specifies which SysRq key functions are enabled by default.
522 This may be set to 1 or 0 to enable or disable them all, or
523 to a bitmask as described in Documentation/admin-guide/sysrq.rst.
525 config MAGIC_SYSRQ_SERIAL
526 bool "Enable magic SysRq key over serial"
527 depends on MAGIC_SYSRQ
530 Many embedded boards have a disconnected TTL level serial which can
531 generate some garbage that can lead to spurious false sysrq detects.
532 This option allows you to decide whether you want to enable the
535 config MAGIC_SYSRQ_SERIAL_SEQUENCE
536 string "Char sequence that enables magic SysRq over serial"
537 depends on MAGIC_SYSRQ_SERIAL
540 Specifies a sequence of characters that can follow BREAK to enable
541 SysRq on a serial console.
543 If unsure, leave an empty string and the option will not be enabled.
546 bool "Debug Filesystem"
548 debugfs is a virtual file system that kernel developers use to put
549 debugging files into. Enable this option to be able to read and
550 write to these files.
552 For detailed documentation on the debugfs API, see
553 Documentation/filesystems/.
558 prompt "Debugfs default access"
560 default DEBUG_FS_ALLOW_ALL
562 This selects the default access restrictions for debugfs.
563 It can be overridden with kernel command line option
564 debugfs=[on,no-mount,off]. The restrictions apply for API access
565 and filesystem registration.
567 config DEBUG_FS_ALLOW_ALL
570 No restrictions apply. Both API and filesystem registration
571 is on. This is the normal default operation.
573 config DEBUG_FS_DISALLOW_MOUNT
574 bool "Do not register debugfs as filesystem"
576 The API is open but filesystem is not loaded. Clients can still do
577 their work and read with debug tools that do not need
580 config DEBUG_FS_ALLOW_NONE
583 Access is off. Clients get -PERM when trying to create nodes in
584 debugfs tree and debugfs is not registered as a filesystem.
585 Client can then back-off or continue without debugfs access.
589 source "lib/Kconfig.kgdb"
590 source "lib/Kconfig.ubsan"
591 source "lib/Kconfig.kcsan"
596 bool "Kernel debugging"
598 Say Y here if you are developing drivers or trying to debug and
599 identify kernel problems.
602 bool "Miscellaneous debug code"
604 depends on DEBUG_KERNEL
606 Say Y here if you need to enable miscellaneous debug code that should
607 be under a more specific debug option but isn't.
610 menu "Memory Debugging"
612 source "mm/Kconfig.debug"
615 bool "Debug object operations"
616 depends on DEBUG_KERNEL
618 If you say Y here, additional code will be inserted into the
619 kernel to track the life time of various objects and validate
620 the operations on those objects.
622 config DEBUG_OBJECTS_SELFTEST
623 bool "Debug objects selftest"
624 depends on DEBUG_OBJECTS
626 This enables the selftest of the object debug code.
628 config DEBUG_OBJECTS_FREE
629 bool "Debug objects in freed memory"
630 depends on DEBUG_OBJECTS
632 This enables checks whether a k/v free operation frees an area
633 which contains an object which has not been deactivated
634 properly. This can make kmalloc/kfree-intensive workloads
637 config DEBUG_OBJECTS_TIMERS
638 bool "Debug timer objects"
639 depends on DEBUG_OBJECTS
641 If you say Y here, additional code will be inserted into the
642 timer routines to track the life time of timer objects and
643 validate the timer operations.
645 config DEBUG_OBJECTS_WORK
646 bool "Debug work objects"
647 depends on DEBUG_OBJECTS
649 If you say Y here, additional code will be inserted into the
650 work queue routines to track the life time of work objects and
651 validate the work operations.
653 config DEBUG_OBJECTS_RCU_HEAD
654 bool "Debug RCU callbacks objects"
655 depends on DEBUG_OBJECTS
657 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
659 config DEBUG_OBJECTS_PERCPU_COUNTER
660 bool "Debug percpu counter objects"
661 depends on DEBUG_OBJECTS
663 If you say Y here, additional code will be inserted into the
664 percpu counter routines to track the life time of percpu counter
665 objects and validate the percpu counter operations.
667 config DEBUG_OBJECTS_ENABLE_DEFAULT
668 int "debug_objects bootup default value (0-1)"
671 depends on DEBUG_OBJECTS
673 Debug objects boot parameter default value
676 bool "Debug slab memory allocations"
677 depends on DEBUG_KERNEL && SLAB
679 Say Y here to have the kernel do limited verification on memory
680 allocation as well as poisoning memory on free to catch use of freed
681 memory. This can make kmalloc/kfree-intensive workloads much slower.
684 bool "SLUB debugging on by default"
685 depends on SLUB && SLUB_DEBUG
688 Boot with debugging on by default. SLUB boots by default with
689 the runtime debug capabilities switched off. Enabling this is
690 equivalent to specifying the "slub_debug" parameter on boot.
691 There is no support for more fine grained debug control like
692 possible with slub_debug=xxx. SLUB debugging may be switched
693 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
698 bool "Enable SLUB performance statistics"
699 depends on SLUB && SYSFS
701 SLUB statistics are useful to debug SLUBs allocation behavior in
702 order find ways to optimize the allocator. This should never be
703 enabled for production use since keeping statistics slows down
704 the allocator by a few percentage points. The slabinfo command
705 supports the determination of the most active slabs to figure
706 out which slabs are relevant to a particular load.
707 Try running: slabinfo -DA
709 config HAVE_DEBUG_KMEMLEAK
712 config DEBUG_KMEMLEAK
713 bool "Kernel memory leak detector"
714 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
716 select STACKTRACE if STACKTRACE_SUPPORT
720 Say Y here if you want to enable the memory leak
721 detector. The memory allocation/freeing is traced in a way
722 similar to the Boehm's conservative garbage collector, the
723 difference being that the orphan objects are not freed but
724 only shown in /sys/kernel/debug/kmemleak. Enabling this
725 feature will introduce an overhead to memory
726 allocations. See Documentation/dev-tools/kmemleak.rst for more
729 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
730 of finding leaks due to the slab objects poisoning.
732 In order to access the kmemleak file, debugfs needs to be
733 mounted (usually at /sys/kernel/debug).
735 config DEBUG_KMEMLEAK_MEM_POOL_SIZE
736 int "Kmemleak memory pool size"
737 depends on DEBUG_KMEMLEAK
741 Kmemleak must track all the memory allocations to avoid
742 reporting false positives. Since memory may be allocated or
743 freed before kmemleak is fully initialised, use a static pool
744 of metadata objects to track such callbacks. After kmemleak is
745 fully initialised, this memory pool acts as an emergency one
746 if slab allocations fail.
748 config DEBUG_KMEMLEAK_TEST
749 tristate "Simple test for the kernel memory leak detector"
750 depends on DEBUG_KMEMLEAK && m
752 This option enables a module that explicitly leaks memory.
756 config DEBUG_KMEMLEAK_DEFAULT_OFF
757 bool "Default kmemleak to off"
758 depends on DEBUG_KMEMLEAK
760 Say Y here to disable kmemleak by default. It can then be enabled
761 on the command line via kmemleak=on.
763 config DEBUG_KMEMLEAK_AUTO_SCAN
764 bool "Enable kmemleak auto scan thread on boot up"
766 depends on DEBUG_KMEMLEAK
768 Depending on the cpu, kmemleak scan may be cpu intensive and can
769 stall user tasks at times. This option enables/disables automatic
770 kmemleak scan at boot up.
772 Say N here to disable kmemleak auto scan thread to stop automatic
773 scanning. Disabling this option disables automatic reporting of
778 config DEBUG_STACK_USAGE
779 bool "Stack utilization instrumentation"
780 depends on DEBUG_KERNEL && !IA64
782 Enables the display of the minimum amount of free stack which each
783 task has ever had available in the sysrq-T and sysrq-P debug output.
785 This option will slow down process creation somewhat.
787 config SCHED_STACK_END_CHECK
788 bool "Detect stack corruption on calls to schedule()"
789 depends on DEBUG_KERNEL
792 This option checks for a stack overrun on calls to schedule().
793 If the stack end location is found to be over written always panic as
794 the content of the corrupted region can no longer be trusted.
795 This is to ensure no erroneous behaviour occurs which could result in
796 data corruption or a sporadic crash at a later stage once the region
797 is examined. The runtime overhead introduced is minimal.
799 config ARCH_HAS_DEBUG_VM_PGTABLE
802 An architecture should select this when it can successfully
803 build and run DEBUG_VM_PGTABLE.
807 depends on DEBUG_KERNEL
809 Enable this to turn on extended checks in the virtual-memory system
810 that may impact performance.
814 config DEBUG_VM_VMACACHE
815 bool "Debug VMA caching"
818 Enable this to turn on VMA caching debug information. Doing so
819 can cause significant overhead, so only enable it in non-production
825 bool "Debug VM red-black trees"
828 Enable VM red-black tree debugging information and extra validations.
832 config DEBUG_VM_PGFLAGS
833 bool "Debug page-flags operations"
836 Enables extra validation on page flags operations.
840 config DEBUG_VM_PGTABLE
841 bool "Debug arch page table for semantics compliance"
843 depends on ARCH_HAS_DEBUG_VM_PGTABLE
844 default y if DEBUG_VM
846 This option provides a debug method which can be used to test
847 architecture page table helper functions on various platforms in
848 verifying if they comply with expected generic MM semantics. This
849 will help architecture code in making sure that any changes or
850 new additions of these helpers still conform to expected
851 semantics of the generic MM. Platforms will have to opt in for
852 this through ARCH_HAS_DEBUG_VM_PGTABLE.
856 config ARCH_HAS_DEBUG_VIRTUAL
860 bool "Debug VM translations"
861 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL
863 Enable some costly sanity checks in virtual to page code. This can
864 catch mistakes with virt_to_page() and friends.
868 config DEBUG_NOMMU_REGIONS
869 bool "Debug the global anon/private NOMMU mapping region tree"
870 depends on DEBUG_KERNEL && !MMU
872 This option causes the global tree of anonymous and private mapping
873 regions to be regularly checked for invalid topology.
875 config DEBUG_MEMORY_INIT
876 bool "Debug memory initialisation" if EXPERT
879 Enable this for additional checks during memory initialisation.
880 The sanity checks verify aspects of the VM such as the memory model
881 and other information provided by the architecture. Verbose
882 information will be printed at KERN_DEBUG loglevel depending
883 on the mminit_loglevel= command-line option.
887 config MEMORY_NOTIFIER_ERROR_INJECT
888 tristate "Memory hotplug notifier error injection module"
889 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION
891 This option provides the ability to inject artificial errors to
892 memory hotplug notifier chain callbacks. It is controlled through
893 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
895 If the notifier call chain should be failed with some events
896 notified, write the error code to "actions/<notifier event>/error".
898 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
900 # cd /sys/kernel/debug/notifier-error-inject/memory
901 # echo -12 > actions/MEM_GOING_OFFLINE/error
902 # echo offline > /sys/devices/system/memory/memoryXXX/state
903 bash: echo: write error: Cannot allocate memory
905 To compile this code as a module, choose M here: the module will
906 be called memory-notifier-error-inject.
910 config DEBUG_PER_CPU_MAPS
911 bool "Debug access to per_cpu maps"
912 depends on DEBUG_KERNEL
915 Say Y to verify that the per_cpu map being accessed has
916 been set up. This adds a fair amount of code to kernel memory
917 and decreases performance.
921 config DEBUG_KMAP_LOCAL
922 bool "Debug kmap_local temporary mappings"
923 depends on DEBUG_KERNEL && KMAP_LOCAL
925 This option enables additional error checking for the kmap_local
926 infrastructure. Disable for production use.
928 config ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
931 config DEBUG_KMAP_LOCAL_FORCE_MAP
932 bool "Enforce kmap_local temporary mappings"
933 depends on DEBUG_KERNEL && ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
935 select DEBUG_KMAP_LOCAL
937 This option enforces temporary mappings through the kmap_local
938 mechanism for non-highmem pages and on non-highmem systems.
939 Disable this for production systems!
942 bool "Highmem debugging"
943 depends on DEBUG_KERNEL && HIGHMEM
944 select DEBUG_KMAP_LOCAL_FORCE_MAP if ARCH_SUPPORTS_KMAP_LOCAL_FORCE_MAP
945 select DEBUG_KMAP_LOCAL
947 This option enables additional error checking for high memory
948 systems. Disable for production systems.
950 config HAVE_DEBUG_STACKOVERFLOW
953 config DEBUG_STACKOVERFLOW
954 bool "Check for stack overflows"
955 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
957 Say Y here if you want to check for overflows of kernel, IRQ
958 and exception stacks (if your architecture uses them). This
959 option will show detailed messages if free stack space drops
960 below a certain limit.
962 These kinds of bugs usually occur when call-chains in the
963 kernel get too deep, especially when interrupts are
966 Use this in cases where you see apparently random memory
967 corruption, especially if it appears in 'struct thread_info'
969 If in doubt, say "N".
971 source "lib/Kconfig.kasan"
972 source "lib/Kconfig.kfence"
974 endmenu # "Memory Debugging"
977 bool "Debug shared IRQ handlers"
978 depends on DEBUG_KERNEL
980 Enable this to generate a spurious interrupt just before a shared
981 interrupt handler is deregistered (generating one when registering
982 is currently disabled). Drivers need to handle this correctly. Some
983 don't and need to be caught.
985 menu "Debug Oops, Lockups and Hangs"
990 Say Y here to enable the kernel to panic when it oopses. This
991 has the same effect as setting oops=panic on the kernel command
994 This feature is useful to ensure that the kernel does not do
995 anything erroneous after an oops which could result in data
996 corruption or other issues.
1000 config PANIC_ON_OOPS_VALUE
1003 default 0 if !PANIC_ON_OOPS
1004 default 1 if PANIC_ON_OOPS
1006 config PANIC_TIMEOUT
1010 Set the timeout value (in seconds) until a reboot occurs when
1011 the kernel panics. If n = 0, then we wait forever. A timeout
1012 value n > 0 will wait n seconds before rebooting, while a timeout
1013 value n < 0 will reboot immediately.
1015 config LOCKUP_DETECTOR
1018 config SOFTLOCKUP_DETECTOR
1019 bool "Detect Soft Lockups"
1020 depends on DEBUG_KERNEL && !S390
1021 select LOCKUP_DETECTOR
1023 Say Y here to enable the kernel to act as a watchdog to detect
1026 Softlockups are bugs that cause the kernel to loop in kernel
1027 mode for more than 20 seconds, without giving other tasks a
1028 chance to run. The current stack trace is displayed upon
1029 detection and the system will stay locked up.
1031 config BOOTPARAM_SOFTLOCKUP_PANIC
1032 bool "Panic (Reboot) On Soft Lockups"
1033 depends on SOFTLOCKUP_DETECTOR
1035 Say Y here to enable the kernel to panic on "soft lockups",
1036 which are bugs that cause the kernel to loop in kernel
1037 mode for more than 20 seconds (configurable using the watchdog_thresh
1038 sysctl), without giving other tasks a chance to run.
1040 The panic can be used in combination with panic_timeout,
1041 to cause the system to reboot automatically after a
1042 lockup has been detected. This feature is useful for
1043 high-availability systems that have uptime guarantees and
1044 where a lockup must be resolved ASAP.
1048 config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
1050 depends on SOFTLOCKUP_DETECTOR
1052 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
1053 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
1055 config HARDLOCKUP_DETECTOR_PERF
1057 select SOFTLOCKUP_DETECTOR
1060 # Enables a timestamp based low pass filter to compensate for perf based
1061 # hard lockup detection which runs too fast due to turbo modes.
1063 config HARDLOCKUP_CHECK_TIMESTAMP
1067 # arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard
1068 # lockup detector rather than the perf based detector.
1070 config HARDLOCKUP_DETECTOR
1071 bool "Detect Hard Lockups"
1072 depends on DEBUG_KERNEL && !S390
1073 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH
1074 select LOCKUP_DETECTOR
1075 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF
1077 Say Y here to enable the kernel to act as a watchdog to detect
1080 Hardlockups are bugs that cause the CPU to loop in kernel mode
1081 for more than 10 seconds, without letting other interrupts have a
1082 chance to run. The current stack trace is displayed upon detection
1083 and the system will stay locked up.
1085 config BOOTPARAM_HARDLOCKUP_PANIC
1086 bool "Panic (Reboot) On Hard Lockups"
1087 depends on HARDLOCKUP_DETECTOR
1089 Say Y here to enable the kernel to panic on "hard lockups",
1090 which are bugs that cause the kernel to loop in kernel
1091 mode with interrupts disabled for more than 10 seconds (configurable
1092 using the watchdog_thresh sysctl).
1096 config BOOTPARAM_HARDLOCKUP_PANIC_VALUE
1098 depends on HARDLOCKUP_DETECTOR
1100 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC
1101 default 1 if BOOTPARAM_HARDLOCKUP_PANIC
1103 config DETECT_HUNG_TASK
1104 bool "Detect Hung Tasks"
1105 depends on DEBUG_KERNEL
1106 default SOFTLOCKUP_DETECTOR
1108 Say Y here to enable the kernel to detect "hung tasks",
1109 which are bugs that cause the task to be stuck in
1110 uninterruptible "D" state indefinitely.
1112 When a hung task is detected, the kernel will print the
1113 current stack trace (which you should report), but the
1114 task will stay in uninterruptible state. If lockdep is
1115 enabled then all held locks will also be reported. This
1116 feature has negligible overhead.
1118 config DEFAULT_HUNG_TASK_TIMEOUT
1119 int "Default timeout for hung task detection (in seconds)"
1120 depends on DETECT_HUNG_TASK
1123 This option controls the default timeout (in seconds) used
1124 to determine when a task has become non-responsive and should
1127 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
1128 sysctl or by writing a value to
1129 /proc/sys/kernel/hung_task_timeout_secs.
1131 A timeout of 0 disables the check. The default is two minutes.
1132 Keeping the default should be fine in most cases.
1134 config BOOTPARAM_HUNG_TASK_PANIC
1135 bool "Panic (Reboot) On Hung Tasks"
1136 depends on DETECT_HUNG_TASK
1138 Say Y here to enable the kernel to panic on "hung tasks",
1139 which are bugs that cause the kernel to leave a task stuck
1140 in uninterruptible "D" state.
1142 The panic can be used in combination with panic_timeout,
1143 to cause the system to reboot automatically after a
1144 hung task has been detected. This feature is useful for
1145 high-availability systems that have uptime guarantees and
1146 where a hung tasks must be resolved ASAP.
1150 config BOOTPARAM_HUNG_TASK_PANIC_VALUE
1152 depends on DETECT_HUNG_TASK
1154 default 0 if !BOOTPARAM_HUNG_TASK_PANIC
1155 default 1 if BOOTPARAM_HUNG_TASK_PANIC
1158 bool "Detect Workqueue Stalls"
1159 depends on DEBUG_KERNEL
1161 Say Y here to enable stall detection on workqueues. If a
1162 worker pool doesn't make forward progress on a pending work
1163 item for over a given amount of time, 30s by default, a
1164 warning message is printed along with dump of workqueue
1165 state. This can be configured through kernel parameter
1166 "workqueue.watchdog_thresh" and its sysfs counterpart.
1169 tristate "Test module to generate lockups"
1172 This builds the "test_lockup" module that helps to make sure
1173 that watchdogs and lockup detectors are working properly.
1175 Depending on module parameters it could emulate soft or hard
1176 lockup, "hung task", or locking arbitrary lock for a long time.
1177 Also it could generate series of lockups with cooling-down periods.
1181 endmenu # "Debug lockups and hangs"
1183 menu "Scheduler Debugging"
1186 bool "Collect scheduler debugging info"
1187 depends on DEBUG_KERNEL && PROC_FS
1190 If you say Y here, the /proc/sched_debug file will be provided
1191 that can help debug the scheduler. The runtime overhead of this
1199 bool "Collect scheduler statistics"
1200 depends on DEBUG_KERNEL && PROC_FS
1203 If you say Y here, additional code will be inserted into the
1204 scheduler and related routines to collect statistics about
1205 scheduler behavior and provide them in /proc/schedstat. These
1206 stats may be useful for both tuning and debugging the scheduler
1207 If you aren't debugging the scheduler or trying to tune a specific
1208 application, you can say N to avoid the very slight overhead
1213 config DEBUG_TIMEKEEPING
1214 bool "Enable extra timekeeping sanity checking"
1216 This option will enable additional timekeeping sanity checks
1217 which may be helpful when diagnosing issues where timekeeping
1218 problems are suspected.
1220 This may include checks in the timekeeping hotpaths, so this
1221 option may have a (very small) performance impact to some
1226 config DEBUG_PREEMPT
1227 bool "Debug preemptible kernel"
1228 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT
1230 If you say Y here then the kernel will use a debug variant of the
1231 commonly used smp_processor_id() function and will print warnings
1232 if kernel code uses it in a preemption-unsafe way. Also, the kernel
1233 will detect preemption count underflows.
1235 This option has potential to introduce high runtime overhead,
1236 depending on workload as it triggers debugging routines for each
1237 this_cpu operation. It should only be used for debugging purposes.
1239 menu "Lock Debugging (spinlocks, mutexes, etc...)"
1241 config LOCK_DEBUGGING_SUPPORT
1243 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1246 config PROVE_LOCKING
1247 bool "Lock debugging: prove locking correctness"
1248 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1250 select DEBUG_SPINLOCK
1251 select DEBUG_MUTEXES if !PREEMPT_RT
1252 select DEBUG_RT_MUTEXES if RT_MUTEXES
1254 select DEBUG_WW_MUTEX_SLOWPATH
1255 select DEBUG_LOCK_ALLOC
1256 select PREEMPT_COUNT if !ARCH_NO_PREEMPT
1257 select TRACE_IRQFLAGS
1260 This feature enables the kernel to prove that all locking
1261 that occurs in the kernel runtime is mathematically
1262 correct: that under no circumstance could an arbitrary (and
1263 not yet triggered) combination of observed locking
1264 sequences (on an arbitrary number of CPUs, running an
1265 arbitrary number of tasks and interrupt contexts) cause a
1268 In short, this feature enables the kernel to report locking
1269 related deadlocks before they actually occur.
1271 The proof does not depend on how hard and complex a
1272 deadlock scenario would be to trigger: how many
1273 participant CPUs, tasks and irq-contexts would be needed
1274 for it to trigger. The proof also does not depend on
1275 timing: if a race and a resulting deadlock is possible
1276 theoretically (no matter how unlikely the race scenario
1277 is), it will be proven so and will immediately be
1278 reported by the kernel (once the event is observed that
1279 makes the deadlock theoretically possible).
1281 If a deadlock is impossible (i.e. the locking rules, as
1282 observed by the kernel, are mathematically correct), the
1283 kernel reports nothing.
1285 NOTE: this feature can also be enabled for rwlocks, mutexes
1286 and rwsems - in which case all dependencies between these
1287 different locking variants are observed and mapped too, and
1288 the proof of observed correctness is also maintained for an
1289 arbitrary combination of these separate locking variants.
1291 For more details, see Documentation/locking/lockdep-design.rst.
1293 config PROVE_RAW_LOCK_NESTING
1294 bool "Enable raw_spinlock - spinlock nesting checks"
1295 depends on PROVE_LOCKING
1298 Enable the raw_spinlock vs. spinlock nesting checks which ensure
1299 that the lock nesting rules for PREEMPT_RT enabled kernels are
1302 NOTE: There are known nesting problems. So if you enable this
1303 option expect lockdep splats until these problems have been fully
1304 addressed which is work in progress. This config switch allows to
1305 identify and analyze these problems. It will be removed and the
1306 check permanently enabled once the main issues have been fixed.
1308 If unsure, select N.
1311 bool "Lock usage statistics"
1312 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1314 select DEBUG_SPINLOCK
1315 select DEBUG_MUTEXES if !PREEMPT_RT
1316 select DEBUG_RT_MUTEXES if RT_MUTEXES
1317 select DEBUG_LOCK_ALLOC
1320 This feature enables tracking lock contention points
1322 For more details, see Documentation/locking/lockstat.rst
1324 This also enables lock events required by "perf lock",
1326 If you want to use "perf lock", you also need to turn on
1327 CONFIG_EVENT_TRACING.
1329 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1330 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1332 config DEBUG_RT_MUTEXES
1333 bool "RT Mutex debugging, deadlock detection"
1334 depends on DEBUG_KERNEL && RT_MUTEXES
1336 This allows rt mutex semantics violations and rt mutex related
1337 deadlocks (lockups) to be detected and reported automatically.
1339 config DEBUG_SPINLOCK
1340 bool "Spinlock and rw-lock debugging: basic checks"
1341 depends on DEBUG_KERNEL
1342 select UNINLINE_SPIN_UNLOCK
1344 Say Y here and build SMP to catch missing spinlock initialization
1345 and certain other kinds of spinlock errors commonly made. This is
1346 best used in conjunction with the NMI watchdog so that spinlock
1347 deadlocks are also debuggable.
1349 config DEBUG_MUTEXES
1350 bool "Mutex debugging: basic checks"
1351 depends on DEBUG_KERNEL && !PREEMPT_RT
1353 This feature allows mutex semantics violations to be detected and
1356 config DEBUG_WW_MUTEX_SLOWPATH
1357 bool "Wait/wound mutex debugging: Slowpath testing"
1358 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1359 select DEBUG_LOCK_ALLOC
1360 select DEBUG_SPINLOCK
1361 select DEBUG_MUTEXES if !PREEMPT_RT
1362 select DEBUG_RT_MUTEXES if PREEMPT_RT
1364 This feature enables slowpath testing for w/w mutex users by
1365 injecting additional -EDEADLK wound/backoff cases. Together with
1366 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1367 will test all possible w/w mutex interface abuse with the
1368 exception of simply not acquiring all the required locks.
1369 Note that this feature can introduce significant overhead, so
1370 it really should not be enabled in a production or distro kernel,
1371 even a debug kernel. If you are a driver writer, enable it. If
1372 you are a distro, do not.
1375 bool "RW Semaphore debugging: basic checks"
1376 depends on DEBUG_KERNEL
1378 This debugging feature allows mismatched rw semaphore locks
1379 and unlocks to be detected and reported.
1381 config DEBUG_LOCK_ALLOC
1382 bool "Lock debugging: detect incorrect freeing of live locks"
1383 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1384 select DEBUG_SPINLOCK
1385 select DEBUG_MUTEXES if !PREEMPT_RT
1386 select DEBUG_RT_MUTEXES if RT_MUTEXES
1389 This feature will check whether any held lock (spinlock, rwlock,
1390 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1391 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1392 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1393 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1394 held during task exit.
1398 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1403 config LOCKDEP_SMALL
1407 int "Bitsize for MAX_LOCKDEP_ENTRIES"
1408 depends on LOCKDEP && !LOCKDEP_SMALL
1412 Try increasing this value if you hit "BUG: MAX_LOCKDEP_ENTRIES too low!" message.
1414 config LOCKDEP_CHAINS_BITS
1415 int "Bitsize for MAX_LOCKDEP_CHAINS"
1416 depends on LOCKDEP && !LOCKDEP_SMALL
1420 Try increasing this value if you hit "BUG: MAX_LOCKDEP_CHAINS too low!" message.
1422 config LOCKDEP_STACK_TRACE_BITS
1423 int "Bitsize for MAX_STACK_TRACE_ENTRIES"
1424 depends on LOCKDEP && !LOCKDEP_SMALL
1428 Try increasing this value if you hit "BUG: MAX_STACK_TRACE_ENTRIES too low!" message.
1430 config LOCKDEP_STACK_TRACE_HASH_BITS
1431 int "Bitsize for STACK_TRACE_HASH_SIZE"
1432 depends on LOCKDEP && !LOCKDEP_SMALL
1436 Try increasing this value if you need large MAX_STACK_TRACE_ENTRIES.
1438 config LOCKDEP_CIRCULAR_QUEUE_BITS
1439 int "Bitsize for elements in circular_queue struct"
1444 Try increasing this value if you hit "lockdep bfs error:-1" warning due to __cq_enqueue() failure.
1446 config DEBUG_LOCKDEP
1447 bool "Lock dependency engine debugging"
1448 depends on DEBUG_KERNEL && LOCKDEP
1449 select DEBUG_IRQFLAGS
1451 If you say Y here, the lock dependency engine will do
1452 additional runtime checks to debug itself, at the price
1453 of more runtime overhead.
1455 config DEBUG_ATOMIC_SLEEP
1456 bool "Sleep inside atomic section checking"
1457 select PREEMPT_COUNT
1458 depends on DEBUG_KERNEL
1459 depends on !ARCH_NO_PREEMPT
1461 If you say Y here, various routines which may sleep will become very
1462 noisy if they are called inside atomic sections: when a spinlock is
1463 held, inside an rcu read side critical section, inside preempt disabled
1464 sections, inside an interrupt, etc...
1466 config DEBUG_LOCKING_API_SELFTESTS
1467 bool "Locking API boot-time self-tests"
1468 depends on DEBUG_KERNEL
1470 Say Y here if you want the kernel to run a short self-test during
1471 bootup. The self-test checks whether common types of locking bugs
1472 are detected by debugging mechanisms or not. (if you disable
1473 lock debugging then those bugs won't be detected of course.)
1474 The following locking APIs are covered: spinlocks, rwlocks,
1477 config LOCK_TORTURE_TEST
1478 tristate "torture tests for locking"
1479 depends on DEBUG_KERNEL
1482 This option provides a kernel module that runs torture tests
1483 on kernel locking primitives. The kernel module may be built
1484 after the fact on the running kernel to be tested, if desired.
1486 Say Y here if you want kernel locking-primitive torture tests
1487 to be built into the kernel.
1488 Say M if you want these torture tests to build as a module.
1489 Say N if you are unsure.
1491 config WW_MUTEX_SELFTEST
1492 tristate "Wait/wound mutex selftests"
1494 This option provides a kernel module that runs tests on the
1495 on the struct ww_mutex locking API.
1497 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction
1498 with this test harness.
1500 Say M if you want these self tests to build as a module.
1501 Say N if you are unsure.
1503 config SCF_TORTURE_TEST
1504 tristate "torture tests for smp_call_function*()"
1505 depends on DEBUG_KERNEL
1508 This option provides a kernel module that runs torture tests
1509 on the smp_call_function() family of primitives. The kernel
1510 module may be built after the fact on the running kernel to
1511 be tested, if desired.
1513 config CSD_LOCK_WAIT_DEBUG
1514 bool "Debugging for csd_lock_wait(), called from smp_call_function*()"
1515 depends on DEBUG_KERNEL
1519 This option enables debug prints when CPUs are slow to respond
1520 to the smp_call_function*() IPI wrappers. These debug prints
1521 include the IPI handler function currently executing (if any)
1522 and relevant stack traces.
1524 endmenu # lock debugging
1526 config TRACE_IRQFLAGS
1527 depends on TRACE_IRQFLAGS_SUPPORT
1530 Enables hooks to interrupt enabling and disabling for
1531 either tracing or lock debugging.
1533 config TRACE_IRQFLAGS_NMI
1535 depends on TRACE_IRQFLAGS
1536 depends on TRACE_IRQFLAGS_NMI_SUPPORT
1538 config DEBUG_IRQFLAGS
1539 bool "Debug IRQ flag manipulation"
1541 Enables checks for potentially unsafe enabling or disabling of
1542 interrupts, such as calling raw_local_irq_restore() when interrupts
1546 bool "Stack backtrace support"
1547 depends on STACKTRACE_SUPPORT
1549 This option causes the kernel to create a /proc/pid/stack for
1550 every process, showing its current stack trace.
1551 It is also used by various kernel debugging features that require
1552 stack trace generation.
1554 config WARN_ALL_UNSEEDED_RANDOM
1555 bool "Warn for all uses of unseeded randomness"
1558 Some parts of the kernel contain bugs relating to their use of
1559 cryptographically secure random numbers before it's actually possible
1560 to generate those numbers securely. This setting ensures that these
1561 flaws don't go unnoticed, by enabling a message, should this ever
1562 occur. This will allow people with obscure setups to know when things
1563 are going wrong, so that they might contact developers about fixing
1566 Unfortunately, on some models of some architectures getting
1567 a fully seeded CRNG is extremely difficult, and so this can
1568 result in dmesg getting spammed for a surprisingly long
1569 time. This is really bad from a security perspective, and
1570 so architecture maintainers really need to do what they can
1571 to get the CRNG seeded sooner after the system is booted.
1572 However, since users cannot do anything actionable to
1573 address this, by default this option is disabled.
1575 Say Y here if you want to receive warnings for all uses of
1576 unseeded randomness. This will be of use primarily for
1577 those developers interested in improving the security of
1578 Linux kernels running on their architecture (or
1581 config DEBUG_KOBJECT
1582 bool "kobject debugging"
1583 depends on DEBUG_KERNEL
1585 If you say Y here, some extra kobject debugging messages will be sent
1588 config DEBUG_KOBJECT_RELEASE
1589 bool "kobject release debugging"
1590 depends on DEBUG_OBJECTS_TIMERS
1592 kobjects are reference counted objects. This means that their
1593 last reference count put is not predictable, and the kobject can
1594 live on past the point at which a driver decides to drop it's
1595 initial reference to the kobject gained on allocation. An
1596 example of this would be a struct device which has just been
1599 However, some buggy drivers assume that after such an operation,
1600 the memory backing the kobject can be immediately freed. This
1601 goes completely against the principles of a refcounted object.
1603 If you say Y here, the kernel will delay the release of kobjects
1604 on the last reference count to improve the visibility of this
1605 kind of kobject release bug.
1607 config HAVE_DEBUG_BUGVERBOSE
1610 menu "Debug kernel data structures"
1613 bool "Debug linked list manipulation"
1614 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION
1616 Enable this to turn on extended checks in the linked-list
1622 bool "Debug priority linked list manipulation"
1623 depends on DEBUG_KERNEL
1625 Enable this to turn on extended checks in the priority-ordered
1626 linked-list (plist) walking routines. This checks the entire
1627 list multiple times during each manipulation.
1632 bool "Debug SG table operations"
1633 depends on DEBUG_KERNEL
1635 Enable this to turn on checks on scatter-gather tables. This can
1636 help find problems with drivers that do not properly initialize
1641 config DEBUG_NOTIFIERS
1642 bool "Debug notifier call chains"
1643 depends on DEBUG_KERNEL
1645 Enable this to turn on sanity checking for notifier call chains.
1646 This is most useful for kernel developers to make sure that
1647 modules properly unregister themselves from notifier chains.
1648 This is a relatively cheap check but if you care about maximum
1651 config BUG_ON_DATA_CORRUPTION
1652 bool "Trigger a BUG when data corruption is detected"
1655 Select this option if the kernel should BUG when it encounters
1656 data corruption in kernel memory structures when they get checked
1663 config DEBUG_CREDENTIALS
1664 bool "Debug credential management"
1665 depends on DEBUG_KERNEL
1667 Enable this to turn on some debug checking for credential
1668 management. The additional code keeps track of the number of
1669 pointers from task_structs to any given cred struct, and checks to
1670 see that this number never exceeds the usage count of the cred
1673 Furthermore, if SELinux is enabled, this also checks that the
1674 security pointer in the cred struct is never seen to be invalid.
1678 source "kernel/rcu/Kconfig.debug"
1680 config DEBUG_WQ_FORCE_RR_CPU
1681 bool "Force round-robin CPU selection for unbound work items"
1682 depends on DEBUG_KERNEL
1685 Workqueue used to implicitly guarantee that work items queued
1686 without explicit CPU specified are put on the local CPU. This
1687 guarantee is no longer true and while local CPU is still
1688 preferred work items may be put on foreign CPUs. Kernel
1689 parameter "workqueue.debug_force_rr_cpu" is added to force
1690 round-robin CPU selection to flush out usages which depend on the
1691 now broken guarantee. This config option enables the debug
1692 feature by default. When enabled, memory and cache locality will
1695 config CPU_HOTPLUG_STATE_CONTROL
1696 bool "Enable CPU hotplug state control"
1697 depends on DEBUG_KERNEL
1698 depends on HOTPLUG_CPU
1701 Allows to write steps between "offline" and "online" to the CPUs
1702 sysfs target file so states can be stepped granular. This is a debug
1703 option for now as the hotplug machinery cannot be stopped and
1704 restarted at arbitrary points yet.
1706 Say N if your are unsure.
1709 bool "Latency measuring infrastructure"
1710 depends on DEBUG_KERNEL
1711 depends on STACKTRACE_SUPPORT
1713 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
1719 Enable this option if you want to use the LatencyTOP tool
1720 to find out which userspace is blocking on what kernel operations.
1722 source "kernel/trace/Kconfig"
1724 config PROVIDE_OHCI1394_DMA_INIT
1725 bool "Remote debugging over FireWire early on boot"
1726 depends on PCI && X86
1728 If you want to debug problems which hang or crash the kernel early
1729 on boot and the crashing machine has a FireWire port, you can use
1730 this feature to remotely access the memory of the crashed machine
1731 over FireWire. This employs remote DMA as part of the OHCI1394
1732 specification which is now the standard for FireWire controllers.
1734 With remote DMA, you can monitor the printk buffer remotely using
1735 firescope and access all memory below 4GB using fireproxy from gdb.
1736 Even controlling a kernel debugger is possible using remote DMA.
1740 If ohci1394_dma=early is used as boot parameter, it will initialize
1741 all OHCI1394 controllers which are found in the PCI config space.
1743 As all changes to the FireWire bus such as enabling and disabling
1744 devices cause a bus reset and thereby disable remote DMA for all
1745 devices, be sure to have the cable plugged and FireWire enabled on
1746 the debugging host before booting the debug target for debugging.
1748 This code (~1k) is freed after boot. By then, the firewire stack
1749 in charge of the OHCI-1394 controllers should be used instead.
1751 See Documentation/core-api/debugging-via-ohci1394.rst for more information.
1753 source "samples/Kconfig"
1755 config ARCH_HAS_DEVMEM_IS_ALLOWED
1758 config STRICT_DEVMEM
1759 bool "Filter access to /dev/mem"
1760 depends on MMU && DEVMEM
1761 depends on ARCH_HAS_DEVMEM_IS_ALLOWED || GENERIC_LIB_DEVMEM_IS_ALLOWED
1762 default y if PPC || X86 || ARM64
1764 If this option is disabled, you allow userspace (root) access to all
1765 of memory, including kernel and userspace memory. Accidental
1766 access to this is obviously disastrous, but specific access can
1767 be used by people debugging the kernel. Note that with PAT support
1768 enabled, even in this case there are restrictions on /dev/mem
1769 use due to the cache aliasing requirements.
1771 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
1772 file only allows userspace access to PCI space and the BIOS code and
1773 data regions. This is sufficient for dosemu and X and all common
1778 config IO_STRICT_DEVMEM
1779 bool "Filter I/O access to /dev/mem"
1780 depends on STRICT_DEVMEM
1782 If this option is disabled, you allow userspace (root) access to all
1783 io-memory regardless of whether a driver is actively using that
1784 range. Accidental access to this is obviously disastrous, but
1785 specific access can be used by people debugging kernel drivers.
1787 If this option is switched on, the /dev/mem file only allows
1788 userspace access to *idle* io-memory ranges (see /proc/iomem) This
1789 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
1790 if the driver using a given range cannot be disabled.
1794 menu "$(SRCARCH) Debugging"
1796 source "arch/$(SRCARCH)/Kconfig.debug"
1800 menu "Kernel Testing and Coverage"
1802 source "lib/kunit/Kconfig"
1804 config NOTIFIER_ERROR_INJECTION
1805 tristate "Notifier error injection"
1806 depends on DEBUG_KERNEL
1809 This option provides the ability to inject artificial errors to
1810 specified notifier chain callbacks. It is useful to test the error
1811 handling of notifier call chain failures.
1815 config PM_NOTIFIER_ERROR_INJECT
1816 tristate "PM notifier error injection module"
1817 depends on PM && NOTIFIER_ERROR_INJECTION
1818 default m if PM_DEBUG
1820 This option provides the ability to inject artificial errors to
1821 PM notifier chain callbacks. It is controlled through debugfs
1822 interface /sys/kernel/debug/notifier-error-inject/pm
1824 If the notifier call chain should be failed with some events
1825 notified, write the error code to "actions/<notifier event>/error".
1827 Example: Inject PM suspend error (-12 = -ENOMEM)
1829 # cd /sys/kernel/debug/notifier-error-inject/pm/
1830 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1831 # echo mem > /sys/power/state
1832 bash: echo: write error: Cannot allocate memory
1834 To compile this code as a module, choose M here: the module will
1835 be called pm-notifier-error-inject.
1839 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1840 tristate "OF reconfig notifier error injection module"
1841 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1843 This option provides the ability to inject artificial errors to
1844 OF reconfig notifier chain callbacks. It is controlled
1845 through debugfs interface under
1846 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1848 If the notifier call chain should be failed with some events
1849 notified, write the error code to "actions/<notifier event>/error".
1851 To compile this code as a module, choose M here: the module will
1852 be called of-reconfig-notifier-error-inject.
1856 config NETDEV_NOTIFIER_ERROR_INJECT
1857 tristate "Netdev notifier error injection module"
1858 depends on NET && NOTIFIER_ERROR_INJECTION
1860 This option provides the ability to inject artificial errors to
1861 netdevice notifier chain callbacks. It is controlled through debugfs
1862 interface /sys/kernel/debug/notifier-error-inject/netdev
1864 If the notifier call chain should be failed with some events
1865 notified, write the error code to "actions/<notifier event>/error".
1867 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1869 # cd /sys/kernel/debug/notifier-error-inject/netdev
1870 # echo -22 > actions/NETDEV_CHANGEMTU/error
1871 # ip link set eth0 mtu 1024
1872 RTNETLINK answers: Invalid argument
1874 To compile this code as a module, choose M here: the module will
1875 be called netdev-notifier-error-inject.
1879 config FUNCTION_ERROR_INJECTION
1880 bool "Fault-injections of functions"
1881 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES
1883 Add fault injections into various functions that are annotated with
1884 ALLOW_ERROR_INJECTION() in the kernel. BPF may also modify the return
1885 value of theses functions. This is useful to test error paths of code.
1889 config FAULT_INJECTION
1890 bool "Fault-injection framework"
1891 depends on DEBUG_KERNEL
1893 Provide fault-injection framework.
1894 For more details, see Documentation/fault-injection/.
1897 bool "Fault-injection capability for kmalloc"
1898 depends on FAULT_INJECTION
1899 depends on SLAB || SLUB
1901 Provide fault-injection capability for kmalloc.
1903 config FAIL_PAGE_ALLOC
1904 bool "Fault-injection capability for alloc_pages()"
1905 depends on FAULT_INJECTION
1907 Provide fault-injection capability for alloc_pages().
1909 config FAULT_INJECTION_USERCOPY
1910 bool "Fault injection capability for usercopy functions"
1911 depends on FAULT_INJECTION
1913 Provides fault-injection capability to inject failures
1914 in usercopy functions (copy_from_user(), get_user(), ...).
1916 config FAIL_MAKE_REQUEST
1917 bool "Fault-injection capability for disk IO"
1918 depends on FAULT_INJECTION && BLOCK
1920 Provide fault-injection capability for disk IO.
1922 config FAIL_IO_TIMEOUT
1923 bool "Fault-injection capability for faking disk interrupts"
1924 depends on FAULT_INJECTION && BLOCK
1926 Provide fault-injection capability on end IO handling. This
1927 will make the block layer "forget" an interrupt as configured,
1928 thus exercising the error handling.
1930 Only works with drivers that use the generic timeout handling,
1931 for others it won't do anything.
1934 bool "Fault-injection capability for futexes"
1936 depends on FAULT_INJECTION && FUTEX
1938 Provide fault-injection capability for futexes.
1940 config FAULT_INJECTION_DEBUG_FS
1941 bool "Debugfs entries for fault-injection capabilities"
1942 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1944 Enable configuration of fault-injection capabilities via debugfs.
1946 config FAIL_FUNCTION
1947 bool "Fault-injection capability for functions"
1948 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION
1950 Provide function-based fault-injection capability.
1951 This will allow you to override a specific function with a return
1952 with given return value. As a result, function caller will see
1953 an error value and have to handle it. This is useful to test the
1954 error handling in various subsystems.
1956 config FAIL_MMC_REQUEST
1957 bool "Fault-injection capability for MMC IO"
1958 depends on FAULT_INJECTION_DEBUG_FS && MMC
1960 Provide fault-injection capability for MMC IO.
1961 This will make the mmc core return data errors. This is
1962 useful to test the error handling in the mmc block device
1963 and to test how the mmc host driver handles retries from
1967 bool "Fault-injection capability for SunRPC"
1968 depends on FAULT_INJECTION_DEBUG_FS && SUNRPC_DEBUG
1970 Provide fault-injection capability for SunRPC and
1973 config FAULT_INJECTION_STACKTRACE_FILTER
1974 bool "stacktrace filter for fault-injection capabilities"
1975 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1978 depends on FRAME_POINTER || MIPS || PPC || S390 || MICROBLAZE || ARM || ARC || X86
1980 Provide stacktrace filter for fault-injection capabilities
1982 config ARCH_HAS_KCOV
1985 An architecture should select this when it can successfully
1986 build and run with CONFIG_KCOV. This typically requires
1987 disabling instrumentation for some early boot code.
1989 config CC_HAS_SANCOV_TRACE_PC
1990 def_bool $(cc-option,-fsanitize-coverage=trace-pc)
1994 bool "Code coverage for fuzzing"
1995 depends on ARCH_HAS_KCOV
1996 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS
1998 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC
2000 KCOV exposes kernel code coverage information in a form suitable
2001 for coverage-guided fuzzing (randomized testing).
2003 If RANDOMIZE_BASE is enabled, PC values will not be stable across
2004 different machines and across reboots. If you need stable PC values,
2005 disable RANDOMIZE_BASE.
2007 For more details, see Documentation/dev-tools/kcov.rst.
2009 config KCOV_ENABLE_COMPARISONS
2010 bool "Enable comparison operands collection by KCOV"
2012 depends on $(cc-option,-fsanitize-coverage=trace-cmp)
2014 KCOV also exposes operands of every comparison in the instrumented
2015 code along with operand sizes and PCs of the comparison instructions.
2016 These operands can be used by fuzzing engines to improve the quality
2017 of fuzzing coverage.
2019 config KCOV_INSTRUMENT_ALL
2020 bool "Instrument all code by default"
2024 If you are doing generic system call fuzzing (like e.g. syzkaller),
2025 then you will want to instrument the whole kernel and you should
2026 say y here. If you are doing more targeted fuzzing (like e.g.
2027 filesystem fuzzing with AFL) then you will want to enable coverage
2028 for more specific subsets of files, and should say n here.
2030 config KCOV_IRQ_AREA_SIZE
2031 hex "Size of interrupt coverage collection area in words"
2035 KCOV uses preallocated per-cpu areas to collect coverage from
2036 soft interrupts. This specifies the size of those areas in the
2037 number of unsigned long words.
2039 menuconfig RUNTIME_TESTING_MENU
2040 bool "Runtime Testing"
2043 if RUNTIME_TESTING_MENU
2046 tristate "Linux Kernel Dump Test Tool Module"
2049 This module enables testing of the different dumping mechanisms by
2050 inducing system failures at predefined crash points.
2051 If you don't need it: say N
2052 Choose M here to compile this code as a module. The module will be
2055 Documentation on how to use the module can be found in
2056 Documentation/fault-injection/provoke-crashes.rst
2058 config TEST_LIST_SORT
2059 tristate "Linked list sorting test" if !KUNIT_ALL_TESTS
2061 default KUNIT_ALL_TESTS
2063 Enable this to turn on 'list_sort()' function test. This test is
2064 executed only once during system boot (so affects only boot time),
2065 or at module load time.
2069 config TEST_MIN_HEAP
2070 tristate "Min heap test"
2071 depends on DEBUG_KERNEL || m
2073 Enable this to turn on min heap function tests. This test is
2074 executed only once during system boot (so affects only boot time),
2075 or at module load time.
2080 tristate "Array-based sort test" if !KUNIT_ALL_TESTS
2082 default KUNIT_ALL_TESTS
2084 This option enables the self-test function of 'sort()' at boot,
2085 or at module load time.
2090 tristate "64bit/32bit division and modulo test"
2091 depends on DEBUG_KERNEL || m
2093 Enable this to turn on 'do_div()' function test. This test is
2094 executed only once during system boot (so affects only boot time),
2095 or at module load time.
2099 config KPROBES_SANITY_TEST
2100 bool "Kprobes sanity tests"
2101 depends on DEBUG_KERNEL
2104 This option provides for testing basic kprobes functionality on
2105 boot. Samples of kprobe and kretprobe are inserted and
2106 verified for functionality.
2108 Say N if you are unsure.
2110 config BACKTRACE_SELF_TEST
2111 tristate "Self test for the backtrace code"
2112 depends on DEBUG_KERNEL
2114 This option provides a kernel module that can be used to test
2115 the kernel stack backtrace code. This option is not useful
2116 for distributions or general kernels, but only for kernel
2117 developers working on architecture code.
2119 Note that if you want to also test saved backtraces, you will
2120 have to enable STACKTRACE as well.
2122 Say N if you are unsure.
2125 tristate "Red-Black tree test"
2126 depends on DEBUG_KERNEL
2128 A benchmark measuring the performance of the rbtree library.
2129 Also includes rbtree invariant checks.
2131 config REED_SOLOMON_TEST
2132 tristate "Reed-Solomon library test"
2133 depends on DEBUG_KERNEL || m
2135 select REED_SOLOMON_ENC16
2136 select REED_SOLOMON_DEC16
2138 This option enables the self-test function of rslib at boot,
2139 or at module load time.
2143 config INTERVAL_TREE_TEST
2144 tristate "Interval tree test"
2145 depends on DEBUG_KERNEL
2146 select INTERVAL_TREE
2148 A benchmark measuring the performance of the interval tree library
2151 tristate "Per cpu operations test"
2152 depends on m && DEBUG_KERNEL
2154 Enable this option to build test module which validates per-cpu
2159 config ATOMIC64_SELFTEST
2160 tristate "Perform an atomic64_t self-test"
2162 Enable this option to test the atomic64_t functions at boot or
2163 at module load time.
2167 config ASYNC_RAID6_TEST
2168 tristate "Self test for hardware accelerated raid6 recovery"
2169 depends on ASYNC_RAID6_RECOV
2172 This is a one-shot self test that permutes through the
2173 recovery of all the possible two disk failure scenarios for a
2174 N-disk array. Recovery is performed with the asynchronous
2175 raid6 recovery routines, and will optionally use an offload
2176 engine if one is available.
2181 tristate "Test functions located in the hexdump module at runtime"
2183 config STRING_SELFTEST
2184 tristate "Test string functions at runtime"
2186 config TEST_STRING_HELPERS
2187 tristate "Test functions located in the string_helpers module at runtime"
2190 tristate "Test strscpy*() family of functions at runtime"
2193 tristate "Test kstrto*() family of functions at runtime"
2196 tristate "Test printf() family of functions at runtime"
2199 tristate "Test scanf() family of functions at runtime"
2202 tristate "Test bitmap_*() family of functions at runtime"
2204 Enable this option to test the bitmap functions at boot.
2209 tristate "Test functions located in the uuid module at runtime"
2212 tristate "Test the XArray code at runtime"
2214 config TEST_OVERFLOW
2215 tristate "Test check_*_overflow() functions at runtime"
2217 config TEST_RHASHTABLE
2218 tristate "Perform selftest on resizable hash table"
2220 Enable this option to test the rhashtable functions at boot.
2225 tristate "Perform selftest on hash functions"
2227 Enable this option to test the kernel's integer (<linux/hash.h>),
2228 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>)
2229 hash functions on boot (or module load).
2231 This is intended to help people writing architecture-specific
2232 optimized versions. If unsure, say N.
2235 tristate "Perform selftest on IDA functions"
2238 tristate "Perform selftest on priority array manager"
2241 Enable this option to test priority array manager on boot
2246 config TEST_IRQ_TIMINGS
2247 bool "IRQ timings selftest"
2248 depends on IRQ_TIMINGS
2250 Enable this option to test the irq timings code on boot.
2255 tristate "Test module loading with 'hello world' module"
2258 This builds the "test_module" module that emits "Hello, world"
2259 on printk when loaded. It is designed to be used for basic
2260 evaluation of the module loading subsystem (for example when
2261 validating module verification). It lacks any extra dependencies,
2262 and will not normally be loaded by the system unless explicitly
2268 tristate "Test module for compilation of bitops operations"
2271 This builds the "test_bitops" module that is much like the
2272 TEST_LKM module except that it does a basic exercise of the
2273 set/clear_bit macros and get_count_order/long to make sure there are
2274 no compiler warnings from C=1 sparse checker or -Wextra
2275 compilations. It has no dependencies and doesn't run or load unless
2276 explicitly requested by name. for example: modprobe test_bitops.
2281 tristate "Test module for stress/performance analysis of vmalloc allocator"
2286 This builds the "test_vmalloc" module that should be used for
2287 stress and performance analysis. So, any new change for vmalloc
2288 subsystem can be evaluated from performance and stability point
2293 config TEST_USER_COPY
2294 tristate "Test user/kernel boundary protections"
2297 This builds the "test_user_copy" module that runs sanity checks
2298 on the copy_to/from_user infrastructure, making sure basic
2299 user/kernel boundary testing is working. If it fails to load,
2300 a regression has been detected in the user/kernel memory boundary
2306 tristate "Test BPF filter functionality"
2309 This builds the "test_bpf" module that runs various test vectors
2310 against the BPF interpreter or BPF JIT compiler depending on the
2311 current setting. This is in particular useful for BPF JIT compiler
2312 development, but also to run regression tests against changes in
2313 the interpreter code. It also enables test stubs for eBPF maps and
2314 verifier used by user space verifier testsuite.
2318 config TEST_BLACKHOLE_DEV
2319 tristate "Test blackhole netdev functionality"
2322 This builds the "test_blackhole_dev" module that validates the
2323 data path through this blackhole netdev.
2327 config FIND_BIT_BENCHMARK
2328 tristate "Test find_bit functions"
2330 This builds the "test_find_bit" module that measure find_*_bit()
2331 functions performance.
2335 config TEST_FIRMWARE
2336 tristate "Test firmware loading via userspace interface"
2337 depends on FW_LOADER
2339 This builds the "test_firmware" module that creates a userspace
2340 interface for testing firmware loading. This can be used to
2341 control the triggering of firmware loading without needing an
2342 actual firmware-using device. The contents can be rechecked by
2348 tristate "sysctl test driver"
2349 depends on PROC_SYSCTL
2351 This builds the "test_sysctl" module. This driver enables to test the
2352 proc sysctl interfaces available to drivers safely without affecting
2353 production knobs which might alter system functionality.
2357 config BITFIELD_KUNIT
2358 tristate "KUnit test bitfield functions at runtime"
2361 Enable this option to test the bitfield functions at boot.
2363 KUnit tests run during boot and output the results to the debug log
2364 in TAP format (http://testanything.org/). Only useful for kernel devs
2365 running the KUnit test harness, and not intended for inclusion into a
2368 For more information on KUnit and unit tests in general please refer
2369 to the KUnit documentation in Documentation/dev-tools/kunit/.
2373 config RESOURCE_KUNIT_TEST
2374 tristate "KUnit test for resource API"
2377 This builds the resource API unit test.
2378 Tests the logic of API provided by resource.c and ioport.h.
2379 For more information on KUnit and unit tests in general please refer
2380 to the KUnit documentation in Documentation/dev-tools/kunit/.
2384 config SYSCTL_KUNIT_TEST
2385 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS
2387 default KUNIT_ALL_TESTS
2389 This builds the proc sysctl unit test, which runs on boot.
2390 Tests the API contract and implementation correctness of sysctl.
2391 For more information on KUnit and unit tests in general please refer
2392 to the KUnit documentation in Documentation/dev-tools/kunit/.
2396 config LIST_KUNIT_TEST
2397 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS
2399 default KUNIT_ALL_TESTS
2401 This builds the linked list KUnit test suite.
2402 It tests that the API and basic functionality of the list_head type
2403 and associated macros.
2405 KUnit tests run during boot and output the results to the debug log
2406 in TAP format (https://testanything.org/). Only useful for kernel devs
2407 running the KUnit test harness, and not intended for inclusion into a
2410 For more information on KUnit and unit tests in general please refer
2411 to the KUnit documentation in Documentation/dev-tools/kunit/.
2415 config LINEAR_RANGES_TEST
2416 tristate "KUnit test for linear_ranges"
2418 select LINEAR_RANGES
2420 This builds the linear_ranges unit test, which runs on boot.
2421 Tests the linear_ranges logic correctness.
2422 For more information on KUnit and unit tests in general please refer
2423 to the KUnit documentation in Documentation/dev-tools/kunit/.
2427 config CMDLINE_KUNIT_TEST
2428 tristate "KUnit test for cmdline API"
2431 This builds the cmdline API unit test.
2432 Tests the logic of API provided by cmdline.c.
2433 For more information on KUnit and unit tests in general please refer
2434 to the KUnit documentation in Documentation/dev-tools/kunit/.
2439 tristate "KUnit test for bits.h"
2442 This builds the bits unit test.
2443 Tests the logic of macros defined in bits.h.
2444 For more information on KUnit and unit tests in general please refer
2445 to the KUnit documentation in Documentation/dev-tools/kunit/.
2449 config SLUB_KUNIT_TEST
2450 tristate "KUnit test for SLUB cache error detection" if !KUNIT_ALL_TESTS
2451 depends on SLUB_DEBUG && KUNIT
2452 default KUNIT_ALL_TESTS
2454 This builds SLUB allocator unit test.
2455 Tests SLUB cache debugging functionality.
2456 For more information on KUnit and unit tests in general please refer
2457 to the KUnit documentation in Documentation/dev-tools/kunit/.
2461 config RATIONAL_KUNIT_TEST
2462 tristate "KUnit test for rational.c" if !KUNIT_ALL_TESTS
2463 depends on KUNIT && RATIONAL
2464 default KUNIT_ALL_TESTS
2466 This builds the rational math unit test.
2467 For more information on KUnit and unit tests in general please refer
2468 to the KUnit documentation in Documentation/dev-tools/kunit/.
2473 tristate "udelay test driver"
2475 This builds the "udelay_test" module that helps to make sure
2476 that udelay() is working properly.
2480 config TEST_STATIC_KEYS
2481 tristate "Test static keys"
2484 Test the static key interfaces.
2489 tristate "kmod stress tester"
2491 depends on NETDEVICES && NET_CORE && INET # for TUN
2498 Test the kernel's module loading mechanism: kmod. kmod implements
2499 support to load modules using the Linux kernel's usermode helper.
2500 This test provides a series of tests against kmod.
2502 Although technically you can either build test_kmod as a module or
2503 into the kernel we disallow building it into the kernel since
2504 it stress tests request_module() and this will very likely cause
2505 some issues by taking over precious threads available from other
2506 module load requests, ultimately this could be fatal.
2510 tools/testing/selftests/kmod/kmod.sh --help
2514 config TEST_DEBUG_VIRTUAL
2515 tristate "Test CONFIG_DEBUG_VIRTUAL feature"
2516 depends on DEBUG_VIRTUAL
2518 Test the kernel's ability to detect incorrect calls to
2519 virt_to_phys() done against the non-linear part of the
2520 kernel's virtual address map.
2524 config TEST_MEMCAT_P
2525 tristate "Test memcat_p() helper function"
2527 Test the memcat_p() helper for correctly merging two
2528 pointer arrays together.
2532 config TEST_LIVEPATCH
2533 tristate "Test livepatching"
2535 depends on DYNAMIC_DEBUG
2536 depends on LIVEPATCH
2539 Test kernel livepatching features for correctness. The tests will
2540 load test modules that will be livepatched in various scenarios.
2542 To run all the livepatching tests:
2544 make -C tools/testing/selftests TARGETS=livepatch run_tests
2546 Alternatively, individual tests may be invoked:
2548 tools/testing/selftests/livepatch/test-callbacks.sh
2549 tools/testing/selftests/livepatch/test-livepatch.sh
2550 tools/testing/selftests/livepatch/test-shadow-vars.sh
2555 tristate "Perform selftest on object aggreration manager"
2559 Enable this option to test object aggregation manager on boot
2563 config TEST_STACKINIT
2564 tristate "Test level of stack variable initialization"
2566 Test if the kernel is zero-initializing stack variables and
2567 padding. Coverage is controlled by compiler flags,
2568 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF,
2569 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL.
2574 tristate "Test heap/page initialization"
2576 Test if the kernel is zero-initializing heap and page allocations.
2577 This can be useful to test init_on_alloc and init_on_free features.
2582 tristate "Test HMM (Heterogeneous Memory Management)"
2583 depends on TRANSPARENT_HUGEPAGE
2584 depends on DEVICE_PRIVATE
2588 This is a pseudo device driver solely for testing HMM.
2589 Say M here if you want to build the HMM test module.
2590 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests.
2594 config TEST_FREE_PAGES
2595 tristate "Test freeing pages"
2597 Test that a memory leak does not occur due to a race between
2598 freeing a block of pages and a speculative page reference.
2599 Loading this module is safe if your kernel has the bug fixed.
2600 If the bug is not fixed, it will leak gigabytes of memory and
2601 probably OOM your system.
2604 tristate "Test floating point operations in kernel space"
2605 depends on X86 && !KCOV_INSTRUMENT_ALL
2607 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu
2608 which will trigger a sequence of floating point operations. This is used
2609 for self-testing floating point control register setting in
2614 config TEST_CLOCKSOURCE_WATCHDOG
2615 tristate "Test clocksource watchdog in kernel space"
2616 depends on CLOCKSOURCE_WATCHDOG
2618 Enable this option to create a kernel module that will trigger
2619 a test of the clocksource watchdog. This module may be loaded
2620 via modprobe or insmod in which case it will run upon being
2621 loaded, or it may be built in, in which case it will run
2626 endif # RUNTIME_TESTING_MENU
2628 config ARCH_USE_MEMTEST
2631 An architecture should select this when it uses early_memtest()
2632 during boot process.
2636 depends on ARCH_USE_MEMTEST
2638 This option adds a kernel parameter 'memtest', which allows memtest
2639 to be set and executed.
2640 memtest=0, mean disabled; -- default
2641 memtest=1, mean do 1 test pattern;
2643 memtest=17, mean do 17 test patterns.
2644 If you are unsure how to answer this question, answer N.
2648 config HYPERV_TESTING
2649 bool "Microsoft Hyper-V driver testing"
2651 depends on HYPERV && DEBUG_FS
2653 Select this option to enable Hyper-V vmbus testing.
2655 endmenu # "Kernel Testing and Coverage"
2657 source "Documentation/Kconfig"
2659 endmenu # Kernel hacking