3 menu "printk and dmesg options"
6 bool "Show timing information on printks"
9 Selecting this option causes time stamps of the printk()
10 messages to be added to the output of the syslog() system
11 call and at the console.
13 The timestamp is always recorded internally, and exported
14 to /dev/kmsg. This flag just specifies if the timestamp should
15 be included, not that the timestamp is recorded.
17 The behavior is also controlled by the kernel command line
18 parameter printk.time=1. See Documentation/admin-guide/kernel-parameters.rst
20 config CONSOLE_LOGLEVEL_DEFAULT
21 int "Default console loglevel (1-15)"
25 Default loglevel to determine what will be printed on the console.
27 Setting a default here is equivalent to passing in loglevel=<x> in
28 the kernel bootargs. loglevel=<x> continues to override whatever
29 value is specified here as well.
31 Note: This does not affect the log level of un-prefixed printk()
32 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT
35 config CONSOLE_LOGLEVEL_QUIET
36 int "quiet console loglevel (1-15)"
40 loglevel to use when "quiet" is passed on the kernel commandline.
42 When "quiet" is passed on the kernel commandline this loglevel
43 will be used as the loglevel. IOW passing "quiet" will be the
44 equivalent of passing "loglevel=<CONSOLE_LOGLEVEL_QUIET>"
46 config MESSAGE_LOGLEVEL_DEFAULT
47 int "Default message log level (1-7)"
51 Default log level for printk statements with no specified priority.
53 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks
54 that are auditing their logs closely may want to set it to a lower
57 Note: This does not affect what message level gets printed on the console
58 by default. To change that, use loglevel=<x> in the kernel bootargs,
59 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value.
61 config BOOT_PRINTK_DELAY
62 bool "Delay each boot printk message by N milliseconds"
63 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY
65 This build option allows you to read kernel boot messages
66 by inserting a short delay after each one. The delay is
67 specified in milliseconds on the kernel command line,
70 It is likely that you would also need to use "lpj=M" to preset
71 the "loops per jiffie" value.
72 See a previous boot log for the "lpj" value to use for your
73 system, and then set "lpj=M" before setting "boot_delay=N".
74 NOTE: Using this option may adversely affect SMP systems.
75 I.e., processors other than the first one may not boot up.
76 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect
77 what it believes to be lockup conditions.
80 bool "Enable dynamic printk() support"
86 Compiles debug level messages into the kernel, which would not
87 otherwise be available at runtime. These messages can then be
88 enabled/disabled based on various levels of scope - per source file,
89 function, module, format string, and line number. This mechanism
90 implicitly compiles in all pr_debug() and dev_dbg() calls, which
91 enlarges the kernel text size by about 2%.
93 If a source file is compiled with DEBUG flag set, any
94 pr_debug() calls in it are enabled by default, but can be
95 disabled at runtime as below. Note that DEBUG flag is
96 turned on by many CONFIG_*DEBUG* options.
100 Dynamic debugging is controlled via the 'dynamic_debug/control' file,
101 which is contained in the 'debugfs' filesystem. Thus, the debugfs
102 filesystem must first be mounted before making use of this feature.
103 We refer the control file as: <debugfs>/dynamic_debug/control. This
104 file contains a list of the debug statements that can be enabled. The
105 format for each line of the file is:
107 filename:lineno [module]function flags format
109 filename : source file of the debug statement
110 lineno : line number of the debug statement
111 module : module that contains the debug statement
112 function : function that contains the debug statement
113 flags : '=p' means the line is turned 'on' for printing
114 format : the format used for the debug statement
118 nullarbor:~ # cat <debugfs>/dynamic_debug/control
119 # filename:lineno [module]function flags format
120 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012"
121 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012"
122 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012"
126 // enable the message at line 1603 of file svcsock.c
127 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' >
128 <debugfs>/dynamic_debug/control
130 // enable all the messages in file svcsock.c
131 nullarbor:~ # echo -n 'file svcsock.c +p' >
132 <debugfs>/dynamic_debug/control
134 // enable all the messages in the NFS server module
135 nullarbor:~ # echo -n 'module nfsd +p' >
136 <debugfs>/dynamic_debug/control
138 // enable all 12 messages in the function svc_process()
139 nullarbor:~ # echo -n 'func svc_process +p' >
140 <debugfs>/dynamic_debug/control
142 // disable all 12 messages in the function svc_process()
143 nullarbor:~ # echo -n 'func svc_process -p' >
144 <debugfs>/dynamic_debug/control
146 See Documentation/admin-guide/dynamic-debug-howto.rst for additional
149 endmenu # "printk and dmesg options"
151 menu "Compile-time checks and compiler options"
154 bool "Compile the kernel with debug info"
155 depends on DEBUG_KERNEL && !COMPILE_TEST
157 If you say Y here the resulting kernel image will include
158 debugging info resulting in a larger kernel image.
159 This adds debug symbols to the kernel and modules (gcc -g), and
160 is needed if you intend to use kernel crashdump or binary object
161 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
162 Say Y here only if you plan to debug the kernel.
166 config DEBUG_INFO_REDUCED
167 bool "Reduce debugging information"
168 depends on DEBUG_INFO
170 If you say Y here gcc is instructed to generate less debugging
171 information for structure types. This means that tools that
172 need full debugging information (like kgdb or systemtap) won't
173 be happy. But if you merely need debugging information to
174 resolve line numbers there is no loss. Advantage is that
175 build directory object sizes shrink dramatically over a full
176 DEBUG_INFO build and compile times are reduced too.
177 Only works with newer gcc versions.
179 config DEBUG_INFO_SPLIT
180 bool "Produce split debuginfo in .dwo files"
181 depends on DEBUG_INFO
183 Generate debug info into separate .dwo files. This significantly
184 reduces the build directory size for builds with DEBUG_INFO,
185 because it stores the information only once on disk in .dwo
186 files instead of multiple times in object files and executables.
187 In addition the debug information is also compressed.
189 Requires recent gcc (4.7+) and recent gdb/binutils.
190 Any tool that packages or reads debug information would need
191 to know about the .dwo files and include them.
192 Incompatible with older versions of ccache.
194 config DEBUG_INFO_DWARF4
195 bool "Generate dwarf4 debuginfo"
196 depends on DEBUG_INFO
198 Generate dwarf4 debug info. This requires recent versions
199 of gcc and gdb. It makes the debug information larger.
200 But it significantly improves the success of resolving
201 variables in gdb on optimized code.
204 bool "Provide GDB scripts for kernel debugging"
205 depends on DEBUG_INFO
207 This creates the required links to GDB helper scripts in the
208 build directory. If you load vmlinux into gdb, the helper
209 scripts will be automatically imported by gdb as well, and
210 additional functions are available to analyze a Linux kernel
211 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst
214 config ENABLE_WARN_DEPRECATED
215 bool "Enable __deprecated logic"
218 Enable the __deprecated logic in the kernel build.
219 Disable this to suppress the "warning: 'foo' is deprecated
220 (declared at kernel/power/somefile.c:1234)" messages.
222 config ENABLE_MUST_CHECK
223 bool "Enable __must_check logic"
226 Enable the __must_check logic in the kernel build. Disable this to
227 suppress the "warning: ignoring return value of 'foo', declared with
228 attribute warn_unused_result" messages.
231 int "Warn for stack frames larger than (needs gcc 4.4)"
233 default 3072 if KASAN_EXTRA
234 default 2048 if GCC_PLUGIN_LATENT_ENTROPY
235 default 1280 if (!64BIT && PARISC)
236 default 1024 if (!64BIT && !PARISC)
237 default 2048 if 64BIT
239 Tell gcc to warn at build time for stack frames larger than this.
240 Setting this too low will cause a lot of warnings.
241 Setting it to 0 disables the warning.
244 config STRIP_ASM_SYMS
245 bool "Strip assembler-generated symbols during link"
248 Strip internal assembler-generated symbols during a link (symbols
249 that look like '.Lxxx') so they don't pollute the output of
250 get_wchan() and suchlike.
253 bool "Generate readable assembler code"
254 depends on DEBUG_KERNEL
256 Disable some compiler optimizations that tend to generate human unreadable
257 assembler output. This may make the kernel slightly slower, but it helps
258 to keep kernel developers who have to stare a lot at assembler listings
261 config UNUSED_SYMBOLS
262 bool "Enable unused/obsolete exported symbols"
265 Unused but exported symbols make the kernel needlessly bigger. For
266 that reason most of these unused exports will soon be removed. This
267 option is provided temporarily to provide a transition period in case
268 some external kernel module needs one of these symbols anyway. If you
269 encounter such a case in your module, consider if you are actually
270 using the right API. (rationale: since nobody in the kernel is using
271 this in a module, there is a pretty good chance it's actually the
272 wrong interface to use). If you really need the symbol, please send a
273 mail to the linux kernel mailing list mentioning the symbol and why
274 you really need it, and what the merge plan to the mainline kernel for
278 bool "Track page owner"
279 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
283 select PAGE_EXTENSION
285 This keeps track of what call chain is the owner of a page, may
286 help to find bare alloc_page(s) leaks. Even if you include this
287 feature on your build, it is disabled in default. You should pass
288 "page_owner=on" to boot parameter in order to enable it. Eats
289 a fair amount of memory if enabled. See tools/vm/page_owner_sort.c
290 for user-space helper.
295 bool "Debug Filesystem"
297 debugfs is a virtual file system that kernel developers use to put
298 debugging files into. Enable this option to be able to read and
299 write to these files.
301 For detailed documentation on the debugfs API, see
302 Documentation/filesystems/.
307 bool "Run 'make headers_check' when building vmlinux"
310 This option will extract the user-visible kernel headers whenever
311 building the kernel, and will run basic sanity checks on them to
312 ensure that exported files do not attempt to include files which
313 were not exported, etc.
315 If you're making modifications to header files which are
316 relevant for userspace, say 'Y', and check the headers
317 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in
318 your build tree), to make sure they're suitable.
320 config DEBUG_SECTION_MISMATCH
321 bool "Enable full Section mismatch analysis"
323 The section mismatch analysis checks if there are illegal
324 references from one section to another section.
325 During linktime or runtime, some sections are dropped;
326 any use of code/data previously in these sections would
327 most likely result in an oops.
328 In the code, functions and variables are annotated with
329 __init,, etc. (see the full list in include/linux/init.h),
330 which results in the code/data being placed in specific sections.
331 The section mismatch analysis is always performed after a full
332 kernel build, and enabling this option causes the following
333 additional steps to occur:
334 - Add the option -fno-inline-functions-called-once to gcc commands.
335 When inlining a function annotated with __init in a non-init
336 function, we would lose the section information and thus
337 the analysis would not catch the illegal reference.
338 This option tells gcc to inline less (but it does result in
340 - Run the section mismatch analysis for each module/built-in.a file.
341 When we run the section mismatch analysis on vmlinux.o, we
342 lose valuable information about where the mismatch was
344 Running the analysis for each module/built-in.a file
345 tells where the mismatch happens much closer to the
346 source. The drawback is that the same mismatch is
347 reported at least twice.
348 - Enable verbose reporting from modpost in order to help resolve
349 the section mismatches that are reported.
351 config SECTION_MISMATCH_WARN_ONLY
352 bool "Make section mismatch errors non-fatal"
355 If you say N here, the build process will fail if there are any
356 section mismatch, instead of just throwing warnings.
361 # Select this config option from the architecture Kconfig, if it
362 # is preferred to always offer frame pointers as a config
363 # option on the architecture (regardless of KERNEL_DEBUG):
365 config ARCH_WANT_FRAME_POINTERS
369 bool "Compile the kernel with frame pointers"
370 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS
371 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
373 If you say Y here the resulting kernel image will be slightly
374 larger and slower, but it gives very useful debugging information
375 in case of kernel bugs. (precise oopses/stacktraces/warnings)
377 config STACK_VALIDATION
378 bool "Compile-time stack metadata validation"
379 depends on HAVE_STACK_VALIDATION
382 Add compile-time checks to validate stack metadata, including frame
383 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure
384 that runtime stack traces are more reliable.
386 This is also a prerequisite for generation of ORC unwind data, which
387 is needed for CONFIG_UNWINDER_ORC.
389 For more information, see
390 tools/objtool/Documentation/stack-validation.txt.
392 config DEBUG_FORCE_WEAK_PER_CPU
393 bool "Force weak per-cpu definitions"
394 depends on DEBUG_KERNEL
396 s390 and alpha require percpu variables in modules to be
397 defined weak to work around addressing range issue which
398 puts the following two restrictions on percpu variable
401 1. percpu symbols must be unique whether static or not
402 2. percpu variables can't be defined inside a function
404 To ensure that generic code follows the above rules, this
405 option forces all percpu variables to be defined as weak.
407 endmenu # "Compiler options"
410 bool "Magic SysRq key"
413 If you say Y here, you will have some control over the system even
414 if the system crashes for example during kernel debugging (e.g., you
415 will be able to flush the buffer cache to disk, reboot the system
416 immediately or dump some status information). This is accomplished
417 by pressing various keys while holding SysRq (Alt+PrintScreen). It
418 also works on a serial console (on PC hardware at least), if you
419 send a BREAK and then within 5 seconds a command keypress. The
420 keys are documented in <file:Documentation/admin-guide/sysrq.rst>.
421 Don't say Y unless you really know what this hack does.
423 config MAGIC_SYSRQ_DEFAULT_ENABLE
424 hex "Enable magic SysRq key functions by default"
425 depends on MAGIC_SYSRQ
428 Specifies which SysRq key functions are enabled by default.
429 This may be set to 1 or 0 to enable or disable them all, or
430 to a bitmask as described in Documentation/admin-guide/sysrq.rst.
432 config MAGIC_SYSRQ_SERIAL
433 bool "Enable magic SysRq key over serial"
434 depends on MAGIC_SYSRQ
437 Many embedded boards have a disconnected TTL level serial which can
438 generate some garbage that can lead to spurious false sysrq detects.
439 This option allows you to decide whether you want to enable the
443 bool "Kernel debugging"
445 Say Y here if you are developing drivers or trying to debug and
446 identify kernel problems.
448 menu "Memory Debugging"
450 source mm/Kconfig.debug
453 bool "Debug object operations"
454 depends on DEBUG_KERNEL
456 If you say Y here, additional code will be inserted into the
457 kernel to track the life time of various objects and validate
458 the operations on those objects.
460 config DEBUG_OBJECTS_SELFTEST
461 bool "Debug objects selftest"
462 depends on DEBUG_OBJECTS
464 This enables the selftest of the object debug code.
466 config DEBUG_OBJECTS_FREE
467 bool "Debug objects in freed memory"
468 depends on DEBUG_OBJECTS
470 This enables checks whether a k/v free operation frees an area
471 which contains an object which has not been deactivated
472 properly. This can make kmalloc/kfree-intensive workloads
475 config DEBUG_OBJECTS_TIMERS
476 bool "Debug timer objects"
477 depends on DEBUG_OBJECTS
479 If you say Y here, additional code will be inserted into the
480 timer routines to track the life time of timer objects and
481 validate the timer operations.
483 config DEBUG_OBJECTS_WORK
484 bool "Debug work objects"
485 depends on DEBUG_OBJECTS
487 If you say Y here, additional code will be inserted into the
488 work queue routines to track the life time of work objects and
489 validate the work operations.
491 config DEBUG_OBJECTS_RCU_HEAD
492 bool "Debug RCU callbacks objects"
493 depends on DEBUG_OBJECTS
495 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
497 config DEBUG_OBJECTS_PERCPU_COUNTER
498 bool "Debug percpu counter objects"
499 depends on DEBUG_OBJECTS
501 If you say Y here, additional code will be inserted into the
502 percpu counter routines to track the life time of percpu counter
503 objects and validate the percpu counter operations.
505 config DEBUG_OBJECTS_ENABLE_DEFAULT
506 int "debug_objects bootup default value (0-1)"
509 depends on DEBUG_OBJECTS
511 Debug objects boot parameter default value
514 bool "Debug slab memory allocations"
515 depends on DEBUG_KERNEL && SLAB
517 Say Y here to have the kernel do limited verification on memory
518 allocation as well as poisoning memory on free to catch use of freed
519 memory. This can make kmalloc/kfree-intensive workloads much slower.
521 config DEBUG_SLAB_LEAK
522 bool "Memory leak debugging"
523 depends on DEBUG_SLAB
526 bool "SLUB debugging on by default"
527 depends on SLUB && SLUB_DEBUG
530 Boot with debugging on by default. SLUB boots by default with
531 the runtime debug capabilities switched off. Enabling this is
532 equivalent to specifying the "slub_debug" parameter on boot.
533 There is no support for more fine grained debug control like
534 possible with slub_debug=xxx. SLUB debugging may be switched
535 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
540 bool "Enable SLUB performance statistics"
541 depends on SLUB && SYSFS
543 SLUB statistics are useful to debug SLUBs allocation behavior in
544 order find ways to optimize the allocator. This should never be
545 enabled for production use since keeping statistics slows down
546 the allocator by a few percentage points. The slabinfo command
547 supports the determination of the most active slabs to figure
548 out which slabs are relevant to a particular load.
549 Try running: slabinfo -DA
551 config HAVE_DEBUG_KMEMLEAK
554 config DEBUG_KMEMLEAK
555 bool "Kernel memory leak detector"
556 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
558 select STACKTRACE if STACKTRACE_SUPPORT
562 Say Y here if you want to enable the memory leak
563 detector. The memory allocation/freeing is traced in a way
564 similar to the Boehm's conservative garbage collector, the
565 difference being that the orphan objects are not freed but
566 only shown in /sys/kernel/debug/kmemleak. Enabling this
567 feature will introduce an overhead to memory
568 allocations. See Documentation/dev-tools/kmemleak.rst for more
571 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
572 of finding leaks due to the slab objects poisoning.
574 In order to access the kmemleak file, debugfs needs to be
575 mounted (usually at /sys/kernel/debug).
577 config DEBUG_KMEMLEAK_EARLY_LOG_SIZE
578 int "Maximum kmemleak early log entries"
579 depends on DEBUG_KMEMLEAK
583 Kmemleak must track all the memory allocations to avoid
584 reporting false positives. Since memory may be allocated or
585 freed before kmemleak is initialised, an early log buffer is
586 used to store these actions. If kmemleak reports "early log
587 buffer exceeded", please increase this value.
589 config DEBUG_KMEMLEAK_TEST
590 tristate "Simple test for the kernel memory leak detector"
591 depends on DEBUG_KMEMLEAK && m
593 This option enables a module that explicitly leaks memory.
597 config DEBUG_KMEMLEAK_DEFAULT_OFF
598 bool "Default kmemleak to off"
599 depends on DEBUG_KMEMLEAK
601 Say Y here to disable kmemleak by default. It can then be enabled
602 on the command line via kmemleak=on.
604 config DEBUG_STACK_USAGE
605 bool "Stack utilization instrumentation"
606 depends on DEBUG_KERNEL && !IA64
608 Enables the display of the minimum amount of free stack which each
609 task has ever had available in the sysrq-T and sysrq-P debug output.
611 This option will slow down process creation somewhat.
615 depends on DEBUG_KERNEL
617 Enable this to turn on extended checks in the virtual-memory system
618 that may impact performance.
622 config DEBUG_VM_VMACACHE
623 bool "Debug VMA caching"
626 Enable this to turn on VMA caching debug information. Doing so
627 can cause significant overhead, so only enable it in non-production
633 bool "Debug VM red-black trees"
636 Enable VM red-black tree debugging information and extra validations.
640 config DEBUG_VM_PGFLAGS
641 bool "Debug page-flags operations"
644 Enables extra validation on page flags operations.
648 config ARCH_HAS_DEBUG_VIRTUAL
652 bool "Debug VM translations"
653 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL
655 Enable some costly sanity checks in virtual to page code. This can
656 catch mistakes with virt_to_page() and friends.
660 config DEBUG_NOMMU_REGIONS
661 bool "Debug the global anon/private NOMMU mapping region tree"
662 depends on DEBUG_KERNEL && !MMU
664 This option causes the global tree of anonymous and private mapping
665 regions to be regularly checked for invalid topology.
667 config DEBUG_MEMORY_INIT
668 bool "Debug memory initialisation" if EXPERT
671 Enable this for additional checks during memory initialisation.
672 The sanity checks verify aspects of the VM such as the memory model
673 and other information provided by the architecture. Verbose
674 information will be printed at KERN_DEBUG loglevel depending
675 on the mminit_loglevel= command-line option.
679 config MEMORY_NOTIFIER_ERROR_INJECT
680 tristate "Memory hotplug notifier error injection module"
681 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION
683 This option provides the ability to inject artificial errors to
684 memory hotplug notifier chain callbacks. It is controlled through
685 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
687 If the notifier call chain should be failed with some events
688 notified, write the error code to "actions/<notifier event>/error".
690 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
692 # cd /sys/kernel/debug/notifier-error-inject/memory
693 # echo -12 > actions/MEM_GOING_OFFLINE/error
694 # echo offline > /sys/devices/system/memory/memoryXXX/state
695 bash: echo: write error: Cannot allocate memory
697 To compile this code as a module, choose M here: the module will
698 be called memory-notifier-error-inject.
702 config DEBUG_PER_CPU_MAPS
703 bool "Debug access to per_cpu maps"
704 depends on DEBUG_KERNEL
707 Say Y to verify that the per_cpu map being accessed has
708 been set up. This adds a fair amount of code to kernel memory
709 and decreases performance.
714 bool "Highmem debugging"
715 depends on DEBUG_KERNEL && HIGHMEM
717 This option enables additional error checking for high memory
718 systems. Disable for production systems.
720 config HAVE_DEBUG_STACKOVERFLOW
723 config DEBUG_STACKOVERFLOW
724 bool "Check for stack overflows"
725 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
727 Say Y here if you want to check for overflows of kernel, IRQ
728 and exception stacks (if your architecture uses them). This
729 option will show detailed messages if free stack space drops
730 below a certain limit.
732 These kinds of bugs usually occur when call-chains in the
733 kernel get too deep, especially when interrupts are
736 Use this in cases where you see apparently random memory
737 corruption, especially if it appears in 'struct thread_info'
739 If in doubt, say "N".
741 source "lib/Kconfig.kasan"
743 endmenu # "Memory Debugging"
748 KCOV does not have any arch-specific code, but currently it is enabled
749 only for x86_64. KCOV requires testing on other archs, and most likely
750 disabling of instrumentation for some early boot code.
752 config CC_HAS_SANCOV_TRACE_PC
753 def_bool $(cc-option,-fsanitize-coverage=trace-pc)
756 bool "Code coverage for fuzzing"
757 depends on ARCH_HAS_KCOV
758 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS
760 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC
762 KCOV exposes kernel code coverage information in a form suitable
763 for coverage-guided fuzzing (randomized testing).
765 If RANDOMIZE_BASE is enabled, PC values will not be stable across
766 different machines and across reboots. If you need stable PC values,
767 disable RANDOMIZE_BASE.
769 For more details, see Documentation/dev-tools/kcov.rst.
771 config KCOV_ENABLE_COMPARISONS
772 bool "Enable comparison operands collection by KCOV"
774 depends on $(cc-option,-fsanitize-coverage=trace-cmp)
776 KCOV also exposes operands of every comparison in the instrumented
777 code along with operand sizes and PCs of the comparison instructions.
778 These operands can be used by fuzzing engines to improve the quality
781 config KCOV_INSTRUMENT_ALL
782 bool "Instrument all code by default"
786 If you are doing generic system call fuzzing (like e.g. syzkaller),
787 then you will want to instrument the whole kernel and you should
788 say y here. If you are doing more targeted fuzzing (like e.g.
789 filesystem fuzzing with AFL) then you will want to enable coverage
790 for more specific subsets of files, and should say n here.
793 bool "Debug shared IRQ handlers"
794 depends on DEBUG_KERNEL
796 Enable this to generate a spurious interrupt as soon as a shared
797 interrupt handler is registered, and just before one is deregistered.
798 Drivers ought to be able to handle interrupts coming in at those
799 points; some don't and need to be caught.
801 menu "Debug Lockups and Hangs"
803 config LOCKUP_DETECTOR
806 config SOFTLOCKUP_DETECTOR
807 bool "Detect Soft Lockups"
808 depends on DEBUG_KERNEL && !S390
809 select LOCKUP_DETECTOR
811 Say Y here to enable the kernel to act as a watchdog to detect
814 Softlockups are bugs that cause the kernel to loop in kernel
815 mode for more than 20 seconds, without giving other tasks a
816 chance to run. The current stack trace is displayed upon
817 detection and the system will stay locked up.
819 config BOOTPARAM_SOFTLOCKUP_PANIC
820 bool "Panic (Reboot) On Soft Lockups"
821 depends on SOFTLOCKUP_DETECTOR
823 Say Y here to enable the kernel to panic on "soft lockups",
824 which are bugs that cause the kernel to loop in kernel
825 mode for more than 20 seconds (configurable using the watchdog_thresh
826 sysctl), without giving other tasks a chance to run.
828 The panic can be used in combination with panic_timeout,
829 to cause the system to reboot automatically after a
830 lockup has been detected. This feature is useful for
831 high-availability systems that have uptime guarantees and
832 where a lockup must be resolved ASAP.
836 config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
838 depends on SOFTLOCKUP_DETECTOR
840 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
841 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
843 config HARDLOCKUP_DETECTOR_PERF
845 select SOFTLOCKUP_DETECTOR
848 # Enables a timestamp based low pass filter to compensate for perf based
849 # hard lockup detection which runs too fast due to turbo modes.
851 config HARDLOCKUP_CHECK_TIMESTAMP
855 # arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard
856 # lockup detector rather than the perf based detector.
858 config HARDLOCKUP_DETECTOR
859 bool "Detect Hard Lockups"
860 depends on DEBUG_KERNEL && !S390
861 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH
862 select LOCKUP_DETECTOR
863 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF
864 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH
866 Say Y here to enable the kernel to act as a watchdog to detect
869 Hardlockups are bugs that cause the CPU to loop in kernel mode
870 for more than 10 seconds, without letting other interrupts have a
871 chance to run. The current stack trace is displayed upon detection
872 and the system will stay locked up.
874 config BOOTPARAM_HARDLOCKUP_PANIC
875 bool "Panic (Reboot) On Hard Lockups"
876 depends on HARDLOCKUP_DETECTOR
878 Say Y here to enable the kernel to panic on "hard lockups",
879 which are bugs that cause the kernel to loop in kernel
880 mode with interrupts disabled for more than 10 seconds (configurable
881 using the watchdog_thresh sysctl).
885 config BOOTPARAM_HARDLOCKUP_PANIC_VALUE
887 depends on HARDLOCKUP_DETECTOR
889 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC
890 default 1 if BOOTPARAM_HARDLOCKUP_PANIC
892 config DETECT_HUNG_TASK
893 bool "Detect Hung Tasks"
894 depends on DEBUG_KERNEL
895 default SOFTLOCKUP_DETECTOR
897 Say Y here to enable the kernel to detect "hung tasks",
898 which are bugs that cause the task to be stuck in
899 uninterruptible "D" state indefinitely.
901 When a hung task is detected, the kernel will print the
902 current stack trace (which you should report), but the
903 task will stay in uninterruptible state. If lockdep is
904 enabled then all held locks will also be reported. This
905 feature has negligible overhead.
907 config DEFAULT_HUNG_TASK_TIMEOUT
908 int "Default timeout for hung task detection (in seconds)"
909 depends on DETECT_HUNG_TASK
912 This option controls the default timeout (in seconds) used
913 to determine when a task has become non-responsive and should
916 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
917 sysctl or by writing a value to
918 /proc/sys/kernel/hung_task_timeout_secs.
920 A timeout of 0 disables the check. The default is two minutes.
921 Keeping the default should be fine in most cases.
923 config BOOTPARAM_HUNG_TASK_PANIC
924 bool "Panic (Reboot) On Hung Tasks"
925 depends on DETECT_HUNG_TASK
927 Say Y here to enable the kernel to panic on "hung tasks",
928 which are bugs that cause the kernel to leave a task stuck
929 in uninterruptible "D" state.
931 The panic can be used in combination with panic_timeout,
932 to cause the system to reboot automatically after a
933 hung task has been detected. This feature is useful for
934 high-availability systems that have uptime guarantees and
935 where a hung tasks must be resolved ASAP.
939 config BOOTPARAM_HUNG_TASK_PANIC_VALUE
941 depends on DETECT_HUNG_TASK
943 default 0 if !BOOTPARAM_HUNG_TASK_PANIC
944 default 1 if BOOTPARAM_HUNG_TASK_PANIC
947 bool "Detect Workqueue Stalls"
948 depends on DEBUG_KERNEL
950 Say Y here to enable stall detection on workqueues. If a
951 worker pool doesn't make forward progress on a pending work
952 item for over a given amount of time, 30s by default, a
953 warning message is printed along with dump of workqueue
954 state. This can be configured through kernel parameter
955 "workqueue.watchdog_thresh" and its sysfs counterpart.
957 endmenu # "Debug lockups and hangs"
962 Say Y here to enable the kernel to panic when it oopses. This
963 has the same effect as setting oops=panic on the kernel command
966 This feature is useful to ensure that the kernel does not do
967 anything erroneous after an oops which could result in data
968 corruption or other issues.
972 config PANIC_ON_OOPS_VALUE
975 default 0 if !PANIC_ON_OOPS
976 default 1 if PANIC_ON_OOPS
982 Set the timeout value (in seconds) until a reboot occurs when the
983 the kernel panics. If n = 0, then we wait forever. A timeout
984 value n > 0 will wait n seconds before rebooting, while a timeout
985 value n < 0 will reboot immediately.
988 bool "Collect scheduler debugging info"
989 depends on DEBUG_KERNEL && PROC_FS
992 If you say Y here, the /proc/sched_debug file will be provided
993 that can help debug the scheduler. The runtime overhead of this
1001 bool "Collect scheduler statistics"
1002 depends on DEBUG_KERNEL && PROC_FS
1005 If you say Y here, additional code will be inserted into the
1006 scheduler and related routines to collect statistics about
1007 scheduler behavior and provide them in /proc/schedstat. These
1008 stats may be useful for both tuning and debugging the scheduler
1009 If you aren't debugging the scheduler or trying to tune a specific
1010 application, you can say N to avoid the very slight overhead
1013 config SCHED_STACK_END_CHECK
1014 bool "Detect stack corruption on calls to schedule()"
1015 depends on DEBUG_KERNEL
1018 This option checks for a stack overrun on calls to schedule().
1019 If the stack end location is found to be over written always panic as
1020 the content of the corrupted region can no longer be trusted.
1021 This is to ensure no erroneous behaviour occurs which could result in
1022 data corruption or a sporadic crash at a later stage once the region
1023 is examined. The runtime overhead introduced is minimal.
1025 config DEBUG_TIMEKEEPING
1026 bool "Enable extra timekeeping sanity checking"
1028 This option will enable additional timekeeping sanity checks
1029 which may be helpful when diagnosing issues where timekeeping
1030 problems are suspected.
1032 This may include checks in the timekeeping hotpaths, so this
1033 option may have a (very small) performance impact to some
1038 config DEBUG_PREEMPT
1039 bool "Debug preemptible kernel"
1040 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT
1043 If you say Y here then the kernel will use a debug variant of the
1044 commonly used smp_processor_id() function and will print warnings
1045 if kernel code uses it in a preemption-unsafe way. Also, the kernel
1046 will detect preemption count underflows.
1048 menu "Lock Debugging (spinlocks, mutexes, etc...)"
1050 config LOCK_DEBUGGING_SUPPORT
1052 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1055 config PROVE_LOCKING
1056 bool "Lock debugging: prove locking correctness"
1057 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1059 select DEBUG_SPINLOCK
1060 select DEBUG_MUTEXES
1061 select DEBUG_RT_MUTEXES if RT_MUTEXES
1062 select DEBUG_RWSEMS if RWSEM_SPIN_ON_OWNER
1063 select DEBUG_WW_MUTEX_SLOWPATH
1064 select DEBUG_LOCK_ALLOC
1065 select TRACE_IRQFLAGS
1068 This feature enables the kernel to prove that all locking
1069 that occurs in the kernel runtime is mathematically
1070 correct: that under no circumstance could an arbitrary (and
1071 not yet triggered) combination of observed locking
1072 sequences (on an arbitrary number of CPUs, running an
1073 arbitrary number of tasks and interrupt contexts) cause a
1076 In short, this feature enables the kernel to report locking
1077 related deadlocks before they actually occur.
1079 The proof does not depend on how hard and complex a
1080 deadlock scenario would be to trigger: how many
1081 participant CPUs, tasks and irq-contexts would be needed
1082 for it to trigger. The proof also does not depend on
1083 timing: if a race and a resulting deadlock is possible
1084 theoretically (no matter how unlikely the race scenario
1085 is), it will be proven so and will immediately be
1086 reported by the kernel (once the event is observed that
1087 makes the deadlock theoretically possible).
1089 If a deadlock is impossible (i.e. the locking rules, as
1090 observed by the kernel, are mathematically correct), the
1091 kernel reports nothing.
1093 NOTE: this feature can also be enabled for rwlocks, mutexes
1094 and rwsems - in which case all dependencies between these
1095 different locking variants are observed and mapped too, and
1096 the proof of observed correctness is also maintained for an
1097 arbitrary combination of these separate locking variants.
1099 For more details, see Documentation/locking/lockdep-design.txt.
1102 bool "Lock usage statistics"
1103 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1105 select DEBUG_SPINLOCK
1106 select DEBUG_MUTEXES
1107 select DEBUG_RT_MUTEXES if RT_MUTEXES
1108 select DEBUG_LOCK_ALLOC
1111 This feature enables tracking lock contention points
1113 For more details, see Documentation/locking/lockstat.txt
1115 This also enables lock events required by "perf lock",
1117 If you want to use "perf lock", you also need to turn on
1118 CONFIG_EVENT_TRACING.
1120 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1121 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1123 config DEBUG_RT_MUTEXES
1124 bool "RT Mutex debugging, deadlock detection"
1125 depends on DEBUG_KERNEL && RT_MUTEXES
1127 This allows rt mutex semantics violations and rt mutex related
1128 deadlocks (lockups) to be detected and reported automatically.
1130 config DEBUG_SPINLOCK
1131 bool "Spinlock and rw-lock debugging: basic checks"
1132 depends on DEBUG_KERNEL
1133 select UNINLINE_SPIN_UNLOCK
1135 Say Y here and build SMP to catch missing spinlock initialization
1136 and certain other kinds of spinlock errors commonly made. This is
1137 best used in conjunction with the NMI watchdog so that spinlock
1138 deadlocks are also debuggable.
1140 config DEBUG_MUTEXES
1141 bool "Mutex debugging: basic checks"
1142 depends on DEBUG_KERNEL
1144 This feature allows mutex semantics violations to be detected and
1147 config DEBUG_WW_MUTEX_SLOWPATH
1148 bool "Wait/wound mutex debugging: Slowpath testing"
1149 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1150 select DEBUG_LOCK_ALLOC
1151 select DEBUG_SPINLOCK
1152 select DEBUG_MUTEXES
1154 This feature enables slowpath testing for w/w mutex users by
1155 injecting additional -EDEADLK wound/backoff cases. Together with
1156 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1157 will test all possible w/w mutex interface abuse with the
1158 exception of simply not acquiring all the required locks.
1159 Note that this feature can introduce significant overhead, so
1160 it really should not be enabled in a production or distro kernel,
1161 even a debug kernel. If you are a driver writer, enable it. If
1162 you are a distro, do not.
1165 bool "RW Semaphore debugging: basic checks"
1166 depends on DEBUG_KERNEL && RWSEM_SPIN_ON_OWNER
1168 This debugging feature allows mismatched rw semaphore locks and unlocks
1169 to be detected and reported.
1171 config DEBUG_LOCK_ALLOC
1172 bool "Lock debugging: detect incorrect freeing of live locks"
1173 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1174 select DEBUG_SPINLOCK
1175 select DEBUG_MUTEXES
1176 select DEBUG_RT_MUTEXES if RT_MUTEXES
1179 This feature will check whether any held lock (spinlock, rwlock,
1180 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1181 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1182 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1183 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1184 held during task exit.
1188 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1190 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE && !ARC && !X86
1194 config LOCKDEP_SMALL
1197 config DEBUG_LOCKDEP
1198 bool "Lock dependency engine debugging"
1199 depends on DEBUG_KERNEL && LOCKDEP
1201 If you say Y here, the lock dependency engine will do
1202 additional runtime checks to debug itself, at the price
1203 of more runtime overhead.
1205 config DEBUG_ATOMIC_SLEEP
1206 bool "Sleep inside atomic section checking"
1207 select PREEMPT_COUNT
1208 depends on DEBUG_KERNEL
1209 depends on !ARCH_NO_PREEMPT
1211 If you say Y here, various routines which may sleep will become very
1212 noisy if they are called inside atomic sections: when a spinlock is
1213 held, inside an rcu read side critical section, inside preempt disabled
1214 sections, inside an interrupt, etc...
1216 config DEBUG_LOCKING_API_SELFTESTS
1217 bool "Locking API boot-time self-tests"
1218 depends on DEBUG_KERNEL
1220 Say Y here if you want the kernel to run a short self-test during
1221 bootup. The self-test checks whether common types of locking bugs
1222 are detected by debugging mechanisms or not. (if you disable
1223 lock debugging then those bugs wont be detected of course.)
1224 The following locking APIs are covered: spinlocks, rwlocks,
1227 config LOCK_TORTURE_TEST
1228 tristate "torture tests for locking"
1229 depends on DEBUG_KERNEL
1233 This option provides a kernel module that runs torture tests
1234 on kernel locking primitives. The kernel module may be built
1235 after the fact on the running kernel to be tested, if desired.
1237 Say Y here if you want kernel locking-primitive torture tests
1238 to be built into the kernel.
1239 Say M if you want these torture tests to build as a module.
1240 Say N if you are unsure.
1242 config WW_MUTEX_SELFTEST
1243 tristate "Wait/wound mutex selftests"
1245 This option provides a kernel module that runs tests on the
1246 on the struct ww_mutex locking API.
1248 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction
1249 with this test harness.
1251 Say M if you want these self tests to build as a module.
1252 Say N if you are unsure.
1254 endmenu # lock debugging
1256 config TRACE_IRQFLAGS
1259 Enables hooks to interrupt enabling and disabling for
1260 either tracing or lock debugging.
1263 bool "Stack backtrace support"
1264 depends on STACKTRACE_SUPPORT
1266 This option causes the kernel to create a /proc/pid/stack for
1267 every process, showing its current stack trace.
1268 It is also used by various kernel debugging features that require
1269 stack trace generation.
1271 config WARN_ALL_UNSEEDED_RANDOM
1272 bool "Warn for all uses of unseeded randomness"
1275 Some parts of the kernel contain bugs relating to their use of
1276 cryptographically secure random numbers before it's actually possible
1277 to generate those numbers securely. This setting ensures that these
1278 flaws don't go unnoticed, by enabling a message, should this ever
1279 occur. This will allow people with obscure setups to know when things
1280 are going wrong, so that they might contact developers about fixing
1283 Unfortunately, on some models of some architectures getting
1284 a fully seeded CRNG is extremely difficult, and so this can
1285 result in dmesg getting spammed for a surprisingly long
1286 time. This is really bad from a security perspective, and
1287 so architecture maintainers really need to do what they can
1288 to get the CRNG seeded sooner after the system is booted.
1289 However, since users can not do anything actionble to
1290 address this, by default the kernel will issue only a single
1291 warning for the first use of unseeded randomness.
1293 Say Y here if you want to receive warnings for all uses of
1294 unseeded randomness. This will be of use primarily for
1295 those developers interersted in improving the security of
1296 Linux kernels running on their architecture (or
1299 config DEBUG_KOBJECT
1300 bool "kobject debugging"
1301 depends on DEBUG_KERNEL
1303 If you say Y here, some extra kobject debugging messages will be sent
1306 config DEBUG_KOBJECT_RELEASE
1307 bool "kobject release debugging"
1308 depends on DEBUG_OBJECTS_TIMERS
1310 kobjects are reference counted objects. This means that their
1311 last reference count put is not predictable, and the kobject can
1312 live on past the point at which a driver decides to drop it's
1313 initial reference to the kobject gained on allocation. An
1314 example of this would be a struct device which has just been
1317 However, some buggy drivers assume that after such an operation,
1318 the memory backing the kobject can be immediately freed. This
1319 goes completely against the principles of a refcounted object.
1321 If you say Y here, the kernel will delay the release of kobjects
1322 on the last reference count to improve the visibility of this
1323 kind of kobject release bug.
1325 config HAVE_DEBUG_BUGVERBOSE
1328 config DEBUG_BUGVERBOSE
1329 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT
1330 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE)
1333 Say Y here to make BUG() panics output the file name and line number
1334 of the BUG call as well as the EIP and oops trace. This aids
1335 debugging but costs about 70-100K of memory.
1338 bool "Debug linked list manipulation"
1339 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION
1341 Enable this to turn on extended checks in the linked-list
1346 config DEBUG_PI_LIST
1347 bool "Debug priority linked list manipulation"
1348 depends on DEBUG_KERNEL
1350 Enable this to turn on extended checks in the priority-ordered
1351 linked-list (plist) walking routines. This checks the entire
1352 list multiple times during each manipulation.
1357 bool "Debug SG table operations"
1358 depends on DEBUG_KERNEL
1360 Enable this to turn on checks on scatter-gather tables. This can
1361 help find problems with drivers that do not properly initialize
1366 config DEBUG_NOTIFIERS
1367 bool "Debug notifier call chains"
1368 depends on DEBUG_KERNEL
1370 Enable this to turn on sanity checking for notifier call chains.
1371 This is most useful for kernel developers to make sure that
1372 modules properly unregister themselves from notifier chains.
1373 This is a relatively cheap check but if you care about maximum
1376 config DEBUG_CREDENTIALS
1377 bool "Debug credential management"
1378 depends on DEBUG_KERNEL
1380 Enable this to turn on some debug checking for credential
1381 management. The additional code keeps track of the number of
1382 pointers from task_structs to any given cred struct, and checks to
1383 see that this number never exceeds the usage count of the cred
1386 Furthermore, if SELinux is enabled, this also checks that the
1387 security pointer in the cred struct is never seen to be invalid.
1391 source "kernel/rcu/Kconfig.debug"
1393 config DEBUG_WQ_FORCE_RR_CPU
1394 bool "Force round-robin CPU selection for unbound work items"
1395 depends on DEBUG_KERNEL
1398 Workqueue used to implicitly guarantee that work items queued
1399 without explicit CPU specified are put on the local CPU. This
1400 guarantee is no longer true and while local CPU is still
1401 preferred work items may be put on foreign CPUs. Kernel
1402 parameter "workqueue.debug_force_rr_cpu" is added to force
1403 round-robin CPU selection to flush out usages which depend on the
1404 now broken guarantee. This config option enables the debug
1405 feature by default. When enabled, memory and cache locality will
1408 config DEBUG_BLOCK_EXT_DEVT
1409 bool "Force extended block device numbers and spread them"
1410 depends on DEBUG_KERNEL
1414 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON
1415 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT
1416 YOU ARE DOING. Distros, please enable this and fix whatever
1419 Conventionally, block device numbers are allocated from
1420 predetermined contiguous area. However, extended block area
1421 may introduce non-contiguous block device numbers. This
1422 option forces most block device numbers to be allocated from
1423 the extended space and spreads them to discover kernel or
1424 userland code paths which assume predetermined contiguous
1425 device number allocation.
1427 Note that turning on this debug option shuffles all the
1428 device numbers for all IDE and SCSI devices including libata
1429 ones, so root partition specified using device number
1430 directly (via rdev or root=MAJ:MIN) won't work anymore.
1431 Textual device names (root=/dev/sdXn) will continue to work.
1433 Say N if you are unsure.
1435 config CPU_HOTPLUG_STATE_CONTROL
1436 bool "Enable CPU hotplug state control"
1437 depends on DEBUG_KERNEL
1438 depends on HOTPLUG_CPU
1441 Allows to write steps between "offline" and "online" to the CPUs
1442 sysfs target file so states can be stepped granular. This is a debug
1443 option for now as the hotplug machinery cannot be stopped and
1444 restarted at arbitrary points yet.
1446 Say N if your are unsure.
1448 config NOTIFIER_ERROR_INJECTION
1449 tristate "Notifier error injection"
1450 depends on DEBUG_KERNEL
1453 This option provides the ability to inject artificial errors to
1454 specified notifier chain callbacks. It is useful to test the error
1455 handling of notifier call chain failures.
1459 config PM_NOTIFIER_ERROR_INJECT
1460 tristate "PM notifier error injection module"
1461 depends on PM && NOTIFIER_ERROR_INJECTION
1462 default m if PM_DEBUG
1464 This option provides the ability to inject artificial errors to
1465 PM notifier chain callbacks. It is controlled through debugfs
1466 interface /sys/kernel/debug/notifier-error-inject/pm
1468 If the notifier call chain should be failed with some events
1469 notified, write the error code to "actions/<notifier event>/error".
1471 Example: Inject PM suspend error (-12 = -ENOMEM)
1473 # cd /sys/kernel/debug/notifier-error-inject/pm/
1474 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1475 # echo mem > /sys/power/state
1476 bash: echo: write error: Cannot allocate memory
1478 To compile this code as a module, choose M here: the module will
1479 be called pm-notifier-error-inject.
1483 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1484 tristate "OF reconfig notifier error injection module"
1485 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1487 This option provides the ability to inject artificial errors to
1488 OF reconfig notifier chain callbacks. It is controlled
1489 through debugfs interface under
1490 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1492 If the notifier call chain should be failed with some events
1493 notified, write the error code to "actions/<notifier event>/error".
1495 To compile this code as a module, choose M here: the module will
1496 be called of-reconfig-notifier-error-inject.
1500 config NETDEV_NOTIFIER_ERROR_INJECT
1501 tristate "Netdev notifier error injection module"
1502 depends on NET && NOTIFIER_ERROR_INJECTION
1504 This option provides the ability to inject artificial errors to
1505 netdevice notifier chain callbacks. It is controlled through debugfs
1506 interface /sys/kernel/debug/notifier-error-inject/netdev
1508 If the notifier call chain should be failed with some events
1509 notified, write the error code to "actions/<notifier event>/error".
1511 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1513 # cd /sys/kernel/debug/notifier-error-inject/netdev
1514 # echo -22 > actions/NETDEV_CHANGEMTU/error
1515 # ip link set eth0 mtu 1024
1516 RTNETLINK answers: Invalid argument
1518 To compile this code as a module, choose M here: the module will
1519 be called netdev-notifier-error-inject.
1523 config FUNCTION_ERROR_INJECTION
1525 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES
1527 config FAULT_INJECTION
1528 bool "Fault-injection framework"
1529 depends on DEBUG_KERNEL
1531 Provide fault-injection framework.
1532 For more details, see Documentation/fault-injection/.
1535 bool "Fault-injection capability for kmalloc"
1536 depends on FAULT_INJECTION
1537 depends on SLAB || SLUB
1539 Provide fault-injection capability for kmalloc.
1541 config FAIL_PAGE_ALLOC
1542 bool "Fault-injection capabilitiy for alloc_pages()"
1543 depends on FAULT_INJECTION
1545 Provide fault-injection capability for alloc_pages().
1547 config FAIL_MAKE_REQUEST
1548 bool "Fault-injection capability for disk IO"
1549 depends on FAULT_INJECTION && BLOCK
1551 Provide fault-injection capability for disk IO.
1553 config FAIL_IO_TIMEOUT
1554 bool "Fault-injection capability for faking disk interrupts"
1555 depends on FAULT_INJECTION && BLOCK
1557 Provide fault-injection capability on end IO handling. This
1558 will make the block layer "forget" an interrupt as configured,
1559 thus exercising the error handling.
1561 Only works with drivers that use the generic timeout handling,
1562 for others it wont do anything.
1565 bool "Fault-injection capability for futexes"
1567 depends on FAULT_INJECTION && FUTEX
1569 Provide fault-injection capability for futexes.
1571 config FAULT_INJECTION_DEBUG_FS
1572 bool "Debugfs entries for fault-injection capabilities"
1573 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1575 Enable configuration of fault-injection capabilities via debugfs.
1577 config FAIL_FUNCTION
1578 bool "Fault-injection capability for functions"
1579 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION
1581 Provide function-based fault-injection capability.
1582 This will allow you to override a specific function with a return
1583 with given return value. As a result, function caller will see
1584 an error value and have to handle it. This is useful to test the
1585 error handling in various subsystems.
1587 config FAIL_MMC_REQUEST
1588 bool "Fault-injection capability for MMC IO"
1589 depends on FAULT_INJECTION_DEBUG_FS && MMC
1591 Provide fault-injection capability for MMC IO.
1592 This will make the mmc core return data errors. This is
1593 useful to test the error handling in the mmc block device
1594 and to test how the mmc host driver handles retries from
1597 config FAULT_INJECTION_STACKTRACE_FILTER
1598 bool "stacktrace filter for fault-injection capabilities"
1599 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1602 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !X86
1604 Provide stacktrace filter for fault-injection capabilities
1607 bool "Latency measuring infrastructure"
1608 depends on DEBUG_KERNEL
1609 depends on STACKTRACE_SUPPORT
1611 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !X86
1618 Enable this option if you want to use the LatencyTOP tool
1619 to find out which userspace is blocking on what kernel operations.
1621 source kernel/trace/Kconfig
1623 config PROVIDE_OHCI1394_DMA_INIT
1624 bool "Remote debugging over FireWire early on boot"
1625 depends on PCI && X86
1627 If you want to debug problems which hang or crash the kernel early
1628 on boot and the crashing machine has a FireWire port, you can use
1629 this feature to remotely access the memory of the crashed machine
1630 over FireWire. This employs remote DMA as part of the OHCI1394
1631 specification which is now the standard for FireWire controllers.
1633 With remote DMA, you can monitor the printk buffer remotely using
1634 firescope and access all memory below 4GB using fireproxy from gdb.
1635 Even controlling a kernel debugger is possible using remote DMA.
1639 If ohci1394_dma=early is used as boot parameter, it will initialize
1640 all OHCI1394 controllers which are found in the PCI config space.
1642 As all changes to the FireWire bus such as enabling and disabling
1643 devices cause a bus reset and thereby disable remote DMA for all
1644 devices, be sure to have the cable plugged and FireWire enabled on
1645 the debugging host before booting the debug target for debugging.
1647 This code (~1k) is freed after boot. By then, the firewire stack
1648 in charge of the OHCI-1394 controllers should be used instead.
1650 See Documentation/debugging-via-ohci1394.txt for more information.
1652 config DMA_API_DEBUG
1653 bool "Enable debugging of DMA-API usage"
1654 select NEED_DMA_MAP_STATE
1656 Enable this option to debug the use of the DMA API by device drivers.
1657 With this option you will be able to detect common bugs in device
1658 drivers like double-freeing of DMA mappings or freeing mappings that
1659 were never allocated.
1661 This also attempts to catch cases where a page owned by DMA is
1662 accessed by the cpu in a way that could cause data corruption. For
1663 example, this enables cow_user_page() to check that the source page is
1666 This option causes a performance degradation. Use only if you want to
1667 debug device drivers and dma interactions.
1671 config DMA_API_DEBUG_SG
1672 bool "Debug DMA scatter-gather usage"
1674 depends on DMA_API_DEBUG
1676 Perform extra checking that callers of dma_map_sg() have respected the
1677 appropriate segment length/boundary limits for the given device when
1678 preparing DMA scatterlists.
1680 This is particularly likely to have been overlooked in cases where the
1681 dma_map_sg() API is used for general bulk mapping of pages rather than
1682 preparing literal scatter-gather descriptors, where there is a risk of
1683 unexpected behaviour from DMA API implementations if the scatterlist
1684 is technically out-of-spec.
1688 menuconfig RUNTIME_TESTING_MENU
1689 bool "Runtime Testing"
1692 if RUNTIME_TESTING_MENU
1695 tristate "Linux Kernel Dump Test Tool Module"
1700 This module enables testing of the different dumping mechanisms by
1701 inducing system failures at predefined crash points.
1702 If you don't need it: say N
1703 Choose M here to compile this code as a module. The module will be
1706 Documentation on how to use the module can be found in
1707 Documentation/fault-injection/provoke-crashes.txt
1709 config TEST_LIST_SORT
1710 tristate "Linked list sorting test"
1711 depends on DEBUG_KERNEL || m
1713 Enable this to turn on 'list_sort()' function test. This test is
1714 executed only once during system boot (so affects only boot time),
1715 or at module load time.
1720 tristate "Array-based sort test"
1721 depends on DEBUG_KERNEL || m
1723 This option enables the self-test function of 'sort()' at boot,
1724 or at module load time.
1728 config KPROBES_SANITY_TEST
1729 bool "Kprobes sanity tests"
1730 depends on DEBUG_KERNEL
1734 This option provides for testing basic kprobes functionality on
1735 boot. Samples of kprobe and kretprobe are inserted and
1736 verified for functionality.
1738 Say N if you are unsure.
1740 config BACKTRACE_SELF_TEST
1741 tristate "Self test for the backtrace code"
1742 depends on DEBUG_KERNEL
1745 This option provides a kernel module that can be used to test
1746 the kernel stack backtrace code. This option is not useful
1747 for distributions or general kernels, but only for kernel
1748 developers working on architecture code.
1750 Note that if you want to also test saved backtraces, you will
1751 have to enable STACKTRACE as well.
1753 Say N if you are unsure.
1756 tristate "Red-Black tree test"
1757 depends on DEBUG_KERNEL
1759 A benchmark measuring the performance of the rbtree library.
1760 Also includes rbtree invariant checks.
1762 config INTERVAL_TREE_TEST
1763 tristate "Interval tree test"
1764 depends on DEBUG_KERNEL
1765 select INTERVAL_TREE
1767 A benchmark measuring the performance of the interval tree library
1770 tristate "Per cpu operations test"
1771 depends on m && DEBUG_KERNEL
1773 Enable this option to build test module which validates per-cpu
1778 config ATOMIC64_SELFTEST
1779 tristate "Perform an atomic64_t self-test"
1781 Enable this option to test the atomic64_t functions at boot or
1782 at module load time.
1786 config ASYNC_RAID6_TEST
1787 tristate "Self test for hardware accelerated raid6 recovery"
1788 depends on ASYNC_RAID6_RECOV
1791 This is a one-shot self test that permutes through the
1792 recovery of all the possible two disk failure scenarios for a
1793 N-disk array. Recovery is performed with the asynchronous
1794 raid6 recovery routines, and will optionally use an offload
1795 engine if one is available.
1800 tristate "Test functions located in the hexdump module at runtime"
1802 config TEST_STRING_HELPERS
1803 tristate "Test functions located in the string_helpers module at runtime"
1806 tristate "Test kstrto*() family of functions at runtime"
1809 tristate "Test printf() family of functions at runtime"
1812 tristate "Test bitmap_*() family of functions at runtime"
1815 Enable this option to test the bitmap functions at boot.
1819 config TEST_BITFIELD
1820 tristate "Test bitfield functions at runtime"
1822 Enable this option to test the bitfield functions at boot.
1827 tristate "Test functions located in the uuid module at runtime"
1829 config TEST_OVERFLOW
1830 tristate "Test check_*_overflow() functions at runtime"
1832 config TEST_RHASHTABLE
1833 tristate "Perform selftest on resizable hash table"
1836 Enable this option to test the rhashtable functions at boot.
1841 tristate "Perform selftest on hash functions"
1844 Enable this option to test the kernel's integer (<linux/hash.h>),
1845 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>)
1846 hash functions on boot (or module load).
1848 This is intended to help people writing architecture-specific
1849 optimized versions. If unsure, say N.
1852 tristate "Perform selftest on priority array manager"
1856 Enable this option to test priority array manager on boot
1862 tristate "Test module loading with 'hello world' module"
1866 This builds the "test_module" module that emits "Hello, world"
1867 on printk when loaded. It is designed to be used for basic
1868 evaluation of the module loading subsystem (for example when
1869 validating module verification). It lacks any extra dependencies,
1870 and will not normally be loaded by the system unless explicitly
1875 config TEST_USER_COPY
1876 tristate "Test user/kernel boundary protections"
1880 This builds the "test_user_copy" module that runs sanity checks
1881 on the copy_to/from_user infrastructure, making sure basic
1882 user/kernel boundary testing is working. If it fails to load,
1883 a regression has been detected in the user/kernel memory boundary
1889 tristate "Test BPF filter functionality"
1893 This builds the "test_bpf" module that runs various test vectors
1894 against the BPF interpreter or BPF JIT compiler depending on the
1895 current setting. This is in particular useful for BPF JIT compiler
1896 development, but also to run regression tests against changes in
1897 the interpreter code. It also enables test stubs for eBPF maps and
1898 verifier used by user space verifier testsuite.
1902 config FIND_BIT_BENCHMARK
1903 tristate "Test find_bit functions"
1906 This builds the "test_find_bit" module that measure find_*_bit()
1907 functions performance.
1911 config TEST_FIRMWARE
1912 tristate "Test firmware loading via userspace interface"
1914 depends on FW_LOADER
1916 This builds the "test_firmware" module that creates a userspace
1917 interface for testing firmware loading. This can be used to
1918 control the triggering of firmware loading without needing an
1919 actual firmware-using device. The contents can be rechecked by
1925 tristate "sysctl test driver"
1927 depends on PROC_SYSCTL
1929 This builds the "test_sysctl" module. This driver enables to test the
1930 proc sysctl interfaces available to drivers safely without affecting
1931 production knobs which might alter system functionality.
1936 tristate "udelay test driver"
1939 This builds the "udelay_test" module that helps to make sure
1940 that udelay() is working properly.
1944 config TEST_STATIC_KEYS
1945 tristate "Test static keys"
1949 Test the static key interfaces.
1954 tristate "kmod stress tester"
1957 depends on BLOCK && (64BIT || LBDAF) # for XFS, BTRFS
1958 depends on NETDEVICES && NET_CORE && INET # for TUN
1964 Test the kernel's module loading mechanism: kmod. kmod implements
1965 support to load modules using the Linux kernel's usermode helper.
1966 This test provides a series of tests against kmod.
1968 Although technically you can either build test_kmod as a module or
1969 into the kernel we disallow building it into the kernel since
1970 it stress tests request_module() and this will very likely cause
1971 some issues by taking over precious threads available from other
1972 module load requests, ultimately this could be fatal.
1976 tools/testing/selftests/kmod/kmod.sh --help
1980 config TEST_DEBUG_VIRTUAL
1981 tristate "Test CONFIG_DEBUG_VIRTUAL feature"
1982 depends on DEBUG_VIRTUAL
1984 Test the kernel's ability to detect incorrect calls to
1985 virt_to_phys() done against the non-linear part of the
1986 kernel's virtual address map.
1990 endif # RUNTIME_TESTING_MENU
1994 depends on HAVE_MEMBLOCK
1996 This option adds a kernel parameter 'memtest', which allows memtest
1998 memtest=0, mean disabled; -- default
1999 memtest=1, mean do 1 test pattern;
2001 memtest=17, mean do 17 test patterns.
2002 If you are unsure how to answer this question, answer N.
2004 config BUG_ON_DATA_CORRUPTION
2005 bool "Trigger a BUG when data corruption is detected"
2008 Select this option if the kernel should BUG when it encounters
2009 data corruption in kernel memory structures when they get checked
2014 source "samples/Kconfig"
2016 source "lib/Kconfig.kgdb"
2018 source "lib/Kconfig.ubsan"
2020 config ARCH_HAS_DEVMEM_IS_ALLOWED
2023 config STRICT_DEVMEM
2024 bool "Filter access to /dev/mem"
2025 depends on MMU && DEVMEM
2026 depends on ARCH_HAS_DEVMEM_IS_ALLOWED
2027 default y if PPC || X86 || ARM64
2029 If this option is disabled, you allow userspace (root) access to all
2030 of memory, including kernel and userspace memory. Accidental
2031 access to this is obviously disastrous, but specific access can
2032 be used by people debugging the kernel. Note that with PAT support
2033 enabled, even in this case there are restrictions on /dev/mem
2034 use due to the cache aliasing requirements.
2036 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
2037 file only allows userspace access to PCI space and the BIOS code and
2038 data regions. This is sufficient for dosemu and X and all common
2043 config IO_STRICT_DEVMEM
2044 bool "Filter I/O access to /dev/mem"
2045 depends on STRICT_DEVMEM
2047 If this option is disabled, you allow userspace (root) access to all
2048 io-memory regardless of whether a driver is actively using that
2049 range. Accidental access to this is obviously disastrous, but
2050 specific access can be used by people debugging kernel drivers.
2052 If this option is switched on, the /dev/mem file only allows
2053 userspace access to *idle* io-memory ranges (see /proc/iomem) This
2054 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
2055 if the driver using a given range cannot be disabled.
2059 source "arch/$(SRCARCH)/Kconfig.debug"
2061 endmenu # Kernel hacking
projects / linux.git / blob