1 menu "printk and dmesg options"
4 bool "Show timing information on printks"
7 Selecting this option causes time stamps of the printk()
8 messages to be added to the output of the syslog() system
9 call and at the console.
11 The timestamp is always recorded internally, and exported
12 to /dev/kmsg. This flag just specifies if the timestamp should
13 be included, not that the timestamp is recorded.
15 The behavior is also controlled by the kernel command line
16 parameter printk.time=1. See Documentation/admin-guide/kernel-parameters.rst
18 config CONSOLE_LOGLEVEL_DEFAULT
19 int "Default console loglevel (1-15)"
23 Default loglevel to determine what will be printed on the console.
25 Setting a default here is equivalent to passing in loglevel=<x> in
26 the kernel bootargs. loglevel=<x> continues to override whatever
27 value is specified here as well.
29 Note: This does not affect the log level of un-prefixed printk()
30 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT
33 config MESSAGE_LOGLEVEL_DEFAULT
34 int "Default message log level (1-7)"
38 Default log level for printk statements with no specified priority.
40 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks
41 that are auditing their logs closely may want to set it to a lower
44 Note: This does not affect what message level gets printed on the console
45 by default. To change that, use loglevel=<x> in the kernel bootargs,
46 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value.
48 config BOOT_PRINTK_DELAY
49 bool "Delay each boot printk message by N milliseconds"
50 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY
52 This build option allows you to read kernel boot messages
53 by inserting a short delay after each one. The delay is
54 specified in milliseconds on the kernel command line,
57 It is likely that you would also need to use "lpj=M" to preset
58 the "loops per jiffie" value.
59 See a previous boot log for the "lpj" value to use for your
60 system, and then set "lpj=M" before setting "boot_delay=N".
61 NOTE: Using this option may adversely affect SMP systems.
62 I.e., processors other than the first one may not boot up.
63 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect
64 what it believes to be lockup conditions.
67 bool "Enable dynamic printk() support"
73 Compiles debug level messages into the kernel, which would not
74 otherwise be available at runtime. These messages can then be
75 enabled/disabled based on various levels of scope - per source file,
76 function, module, format string, and line number. This mechanism
77 implicitly compiles in all pr_debug() and dev_dbg() calls, which
78 enlarges the kernel text size by about 2%.
80 If a source file is compiled with DEBUG flag set, any
81 pr_debug() calls in it are enabled by default, but can be
82 disabled at runtime as below. Note that DEBUG flag is
83 turned on by many CONFIG_*DEBUG* options.
87 Dynamic debugging is controlled via the 'dynamic_debug/control' file,
88 which is contained in the 'debugfs' filesystem. Thus, the debugfs
89 filesystem must first be mounted before making use of this feature.
90 We refer the control file as: <debugfs>/dynamic_debug/control. This
91 file contains a list of the debug statements that can be enabled. The
92 format for each line of the file is:
94 filename:lineno [module]function flags format
96 filename : source file of the debug statement
97 lineno : line number of the debug statement
98 module : module that contains the debug statement
99 function : function that contains the debug statement
100 flags : '=p' means the line is turned 'on' for printing
101 format : the format used for the debug statement
105 nullarbor:~ # cat <debugfs>/dynamic_debug/control
106 # filename:lineno [module]function flags format
107 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012"
108 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012"
109 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012"
113 // enable the message at line 1603 of file svcsock.c
114 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' >
115 <debugfs>/dynamic_debug/control
117 // enable all the messages in file svcsock.c
118 nullarbor:~ # echo -n 'file svcsock.c +p' >
119 <debugfs>/dynamic_debug/control
121 // enable all the messages in the NFS server module
122 nullarbor:~ # echo -n 'module nfsd +p' >
123 <debugfs>/dynamic_debug/control
125 // enable all 12 messages in the function svc_process()
126 nullarbor:~ # echo -n 'func svc_process +p' >
127 <debugfs>/dynamic_debug/control
129 // disable all 12 messages in the function svc_process()
130 nullarbor:~ # echo -n 'func svc_process -p' >
131 <debugfs>/dynamic_debug/control
133 See Documentation/admin-guide/dynamic-debug-howto.rst for additional
136 endmenu # "printk and dmesg options"
138 menu "Compile-time checks and compiler options"
141 bool "Compile the kernel with debug info"
142 depends on DEBUG_KERNEL && !COMPILE_TEST
144 If you say Y here the resulting kernel image will include
145 debugging info resulting in a larger kernel image.
146 This adds debug symbols to the kernel and modules (gcc -g), and
147 is needed if you intend to use kernel crashdump or binary object
148 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
149 Say Y here only if you plan to debug the kernel.
153 config DEBUG_INFO_REDUCED
154 bool "Reduce debugging information"
155 depends on DEBUG_INFO
157 If you say Y here gcc is instructed to generate less debugging
158 information for structure types. This means that tools that
159 need full debugging information (like kgdb or systemtap) won't
160 be happy. But if you merely need debugging information to
161 resolve line numbers there is no loss. Advantage is that
162 build directory object sizes shrink dramatically over a full
163 DEBUG_INFO build and compile times are reduced too.
164 Only works with newer gcc versions.
166 config DEBUG_INFO_SPLIT
167 bool "Produce split debuginfo in .dwo files"
168 depends on DEBUG_INFO && !FRV
170 Generate debug info into separate .dwo files. This significantly
171 reduces the build directory size for builds with DEBUG_INFO,
172 because it stores the information only once on disk in .dwo
173 files instead of multiple times in object files and executables.
174 In addition the debug information is also compressed.
176 Requires recent gcc (4.7+) and recent gdb/binutils.
177 Any tool that packages or reads debug information would need
178 to know about the .dwo files and include them.
179 Incompatible with older versions of ccache.
181 config DEBUG_INFO_DWARF4
182 bool "Generate dwarf4 debuginfo"
183 depends on DEBUG_INFO
185 Generate dwarf4 debug info. This requires recent versions
186 of gcc and gdb. It makes the debug information larger.
187 But it significantly improves the success of resolving
188 variables in gdb on optimized code.
191 bool "Provide GDB scripts for kernel debugging"
192 depends on DEBUG_INFO
194 This creates the required links to GDB helper scripts in the
195 build directory. If you load vmlinux into gdb, the helper
196 scripts will be automatically imported by gdb as well, and
197 additional functions are available to analyze a Linux kernel
198 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst
201 config ENABLE_WARN_DEPRECATED
202 bool "Enable __deprecated logic"
205 Enable the __deprecated logic in the kernel build.
206 Disable this to suppress the "warning: 'foo' is deprecated
207 (declared at kernel/power/somefile.c:1234)" messages.
209 config ENABLE_MUST_CHECK
210 bool "Enable __must_check logic"
213 Enable the __must_check logic in the kernel build. Disable this to
214 suppress the "warning: ignoring return value of 'foo', declared with
215 attribute warn_unused_result" messages.
218 int "Warn for stack frames larger than (needs gcc 4.4)"
221 default 2048 if GCC_PLUGIN_LATENT_ENTROPY
222 default 1024 if !64BIT
223 default 2048 if 64BIT
225 Tell gcc to warn at build time for stack frames larger than this.
226 Setting this too low will cause a lot of warnings.
227 Setting it to 0 disables the warning.
230 config STRIP_ASM_SYMS
231 bool "Strip assembler-generated symbols during link"
234 Strip internal assembler-generated symbols during a link (symbols
235 that look like '.Lxxx') so they don't pollute the output of
236 get_wchan() and suchlike.
239 bool "Generate readable assembler code"
240 depends on DEBUG_KERNEL
242 Disable some compiler optimizations that tend to generate human unreadable
243 assembler output. This may make the kernel slightly slower, but it helps
244 to keep kernel developers who have to stare a lot at assembler listings
247 config UNUSED_SYMBOLS
248 bool "Enable unused/obsolete exported symbols"
251 Unused but exported symbols make the kernel needlessly bigger. For
252 that reason most of these unused exports will soon be removed. This
253 option is provided temporarily to provide a transition period in case
254 some external kernel module needs one of these symbols anyway. If you
255 encounter such a case in your module, consider if you are actually
256 using the right API. (rationale: since nobody in the kernel is using
257 this in a module, there is a pretty good chance it's actually the
258 wrong interface to use). If you really need the symbol, please send a
259 mail to the linux kernel mailing list mentioning the symbol and why
260 you really need it, and what the merge plan to the mainline kernel for
264 bool "Track page owner"
265 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
269 select PAGE_EXTENSION
271 This keeps track of what call chain is the owner of a page, may
272 help to find bare alloc_page(s) leaks. Even if you include this
273 feature on your build, it is disabled in default. You should pass
274 "page_owner=on" to boot parameter in order to enable it. Eats
275 a fair amount of memory if enabled. See tools/vm/page_owner_sort.c
276 for user-space helper.
281 bool "Debug Filesystem"
284 debugfs is a virtual file system that kernel developers use to put
285 debugging files into. Enable this option to be able to read and
286 write to these files.
288 For detailed documentation on the debugfs API, see
289 Documentation/DocBook/filesystems.
294 bool "Run 'make headers_check' when building vmlinux"
297 This option will extract the user-visible kernel headers whenever
298 building the kernel, and will run basic sanity checks on them to
299 ensure that exported files do not attempt to include files which
300 were not exported, etc.
302 If you're making modifications to header files which are
303 relevant for userspace, say 'Y', and check the headers
304 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in
305 your build tree), to make sure they're suitable.
307 config DEBUG_SECTION_MISMATCH
308 bool "Enable full Section mismatch analysis"
310 The section mismatch analysis checks if there are illegal
311 references from one section to another section.
312 During linktime or runtime, some sections are dropped;
313 any use of code/data previously in these sections would
314 most likely result in an oops.
315 In the code, functions and variables are annotated with
316 __init,, etc. (see the full list in include/linux/init.h),
317 which results in the code/data being placed in specific sections.
318 The section mismatch analysis is always performed after a full
319 kernel build, and enabling this option causes the following
320 additional steps to occur:
321 - Add the option -fno-inline-functions-called-once to gcc commands.
322 When inlining a function annotated with __init in a non-init
323 function, we would lose the section information and thus
324 the analysis would not catch the illegal reference.
325 This option tells gcc to inline less (but it does result in
327 - Run the section mismatch analysis for each module/built-in.o file.
328 When we run the section mismatch analysis on vmlinux.o, we
329 lose valuable information about where the mismatch was
331 Running the analysis for each module/built-in.o file
332 tells where the mismatch happens much closer to the
333 source. The drawback is that the same mismatch is
334 reported at least twice.
335 - Enable verbose reporting from modpost in order to help resolve
336 the section mismatches that are reported.
338 config SECTION_MISMATCH_WARN_ONLY
339 bool "Make section mismatch errors non-fatal"
342 If you say N here, the build process will fail if there are any
343 section mismatch, instead of just throwing warnings.
348 # Select this config option from the architecture Kconfig, if it
349 # is preferred to always offer frame pointers as a config
350 # option on the architecture (regardless of KERNEL_DEBUG):
352 config ARCH_WANT_FRAME_POINTERS
357 bool "Compile the kernel with frame pointers"
358 depends on DEBUG_KERNEL && \
359 (CRIS || M68K || FRV || UML || \
360 SUPERH || BLACKFIN || MN10300 || METAG) || \
361 ARCH_WANT_FRAME_POINTERS
362 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
364 If you say Y here the resulting kernel image will be slightly
365 larger and slower, but it gives very useful debugging information
366 in case of kernel bugs. (precise oopses/stacktraces/warnings)
368 config STACK_VALIDATION
369 bool "Compile-time stack metadata validation"
370 depends on HAVE_STACK_VALIDATION
373 Add compile-time checks to validate stack metadata, including frame
374 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure
375 that runtime stack traces are more reliable.
377 For more information, see
378 tools/objtool/Documentation/stack-validation.txt.
380 config DEBUG_FORCE_WEAK_PER_CPU
381 bool "Force weak per-cpu definitions"
382 depends on DEBUG_KERNEL
384 s390 and alpha require percpu variables in modules to be
385 defined weak to work around addressing range issue which
386 puts the following two restrictions on percpu variable
389 1. percpu symbols must be unique whether static or not
390 2. percpu variables can't be defined inside a function
392 To ensure that generic code follows the above rules, this
393 option forces all percpu variables to be defined as weak.
395 endmenu # "Compiler options"
398 bool "Magic SysRq key"
401 If you say Y here, you will have some control over the system even
402 if the system crashes for example during kernel debugging (e.g., you
403 will be able to flush the buffer cache to disk, reboot the system
404 immediately or dump some status information). This is accomplished
405 by pressing various keys while holding SysRq (Alt+PrintScreen). It
406 also works on a serial console (on PC hardware at least), if you
407 send a BREAK and then within 5 seconds a command keypress. The
408 keys are documented in <file:Documentation/admin-guide/sysrq.rst>.
409 Don't say Y unless you really know what this hack does.
411 config MAGIC_SYSRQ_DEFAULT_ENABLE
412 hex "Enable magic SysRq key functions by default"
413 depends on MAGIC_SYSRQ
416 Specifies which SysRq key functions are enabled by default.
417 This may be set to 1 or 0 to enable or disable them all, or
418 to a bitmask as described in Documentation/admin-guide/sysrq.rst.
420 config MAGIC_SYSRQ_SERIAL
421 bool "Enable magic SysRq key over serial"
422 depends on MAGIC_SYSRQ
425 Many embedded boards have a disconnected TTL level serial which can
426 generate some garbage that can lead to spurious false sysrq detects.
427 This option allows you to decide whether you want to enable the
431 bool "Kernel debugging"
433 Say Y here if you are developing drivers or trying to debug and
434 identify kernel problems.
436 menu "Memory Debugging"
438 source mm/Kconfig.debug
441 bool "Debug object operations"
442 depends on DEBUG_KERNEL
444 If you say Y here, additional code will be inserted into the
445 kernel to track the life time of various objects and validate
446 the operations on those objects.
448 config DEBUG_OBJECTS_SELFTEST
449 bool "Debug objects selftest"
450 depends on DEBUG_OBJECTS
452 This enables the selftest of the object debug code.
454 config DEBUG_OBJECTS_FREE
455 bool "Debug objects in freed memory"
456 depends on DEBUG_OBJECTS
458 This enables checks whether a k/v free operation frees an area
459 which contains an object which has not been deactivated
460 properly. This can make kmalloc/kfree-intensive workloads
463 config DEBUG_OBJECTS_TIMERS
464 bool "Debug timer objects"
465 depends on DEBUG_OBJECTS
467 If you say Y here, additional code will be inserted into the
468 timer routines to track the life time of timer objects and
469 validate the timer operations.
471 config DEBUG_OBJECTS_WORK
472 bool "Debug work objects"
473 depends on DEBUG_OBJECTS
475 If you say Y here, additional code will be inserted into the
476 work queue routines to track the life time of work objects and
477 validate the work operations.
479 config DEBUG_OBJECTS_RCU_HEAD
480 bool "Debug RCU callbacks objects"
481 depends on DEBUG_OBJECTS
483 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
485 config DEBUG_OBJECTS_PERCPU_COUNTER
486 bool "Debug percpu counter objects"
487 depends on DEBUG_OBJECTS
489 If you say Y here, additional code will be inserted into the
490 percpu counter routines to track the life time of percpu counter
491 objects and validate the percpu counter operations.
493 config DEBUG_OBJECTS_ENABLE_DEFAULT
494 int "debug_objects bootup default value (0-1)"
497 depends on DEBUG_OBJECTS
499 Debug objects boot parameter default value
502 bool "Debug slab memory allocations"
503 depends on DEBUG_KERNEL && SLAB && !KMEMCHECK
505 Say Y here to have the kernel do limited verification on memory
506 allocation as well as poisoning memory on free to catch use of freed
507 memory. This can make kmalloc/kfree-intensive workloads much slower.
509 config DEBUG_SLAB_LEAK
510 bool "Memory leak debugging"
511 depends on DEBUG_SLAB
514 bool "SLUB debugging on by default"
515 depends on SLUB && SLUB_DEBUG && !KMEMCHECK
518 Boot with debugging on by default. SLUB boots by default with
519 the runtime debug capabilities switched off. Enabling this is
520 equivalent to specifying the "slub_debug" parameter on boot.
521 There is no support for more fine grained debug control like
522 possible with slub_debug=xxx. SLUB debugging may be switched
523 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
528 bool "Enable SLUB performance statistics"
529 depends on SLUB && SYSFS
531 SLUB statistics are useful to debug SLUBs allocation behavior in
532 order find ways to optimize the allocator. This should never be
533 enabled for production use since keeping statistics slows down
534 the allocator by a few percentage points. The slabinfo command
535 supports the determination of the most active slabs to figure
536 out which slabs are relevant to a particular load.
537 Try running: slabinfo -DA
539 config HAVE_DEBUG_KMEMLEAK
542 config DEBUG_KMEMLEAK
543 bool "Kernel memory leak detector"
544 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
546 select STACKTRACE if STACKTRACE_SUPPORT
550 Say Y here if you want to enable the memory leak
551 detector. The memory allocation/freeing is traced in a way
552 similar to the Boehm's conservative garbage collector, the
553 difference being that the orphan objects are not freed but
554 only shown in /sys/kernel/debug/kmemleak. Enabling this
555 feature will introduce an overhead to memory
556 allocations. See Documentation/dev-tools/kmemleak.rst for more
559 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
560 of finding leaks due to the slab objects poisoning.
562 In order to access the kmemleak file, debugfs needs to be
563 mounted (usually at /sys/kernel/debug).
565 config DEBUG_KMEMLEAK_EARLY_LOG_SIZE
566 int "Maximum kmemleak early log entries"
567 depends on DEBUG_KMEMLEAK
571 Kmemleak must track all the memory allocations to avoid
572 reporting false positives. Since memory may be allocated or
573 freed before kmemleak is initialised, an early log buffer is
574 used to store these actions. If kmemleak reports "early log
575 buffer exceeded", please increase this value.
577 config DEBUG_KMEMLEAK_TEST
578 tristate "Simple test for the kernel memory leak detector"
579 depends on DEBUG_KMEMLEAK && m
581 This option enables a module that explicitly leaks memory.
585 config DEBUG_KMEMLEAK_DEFAULT_OFF
586 bool "Default kmemleak to off"
587 depends on DEBUG_KMEMLEAK
589 Say Y here to disable kmemleak by default. It can then be enabled
590 on the command line via kmemleak=on.
592 config DEBUG_STACK_USAGE
593 bool "Stack utilization instrumentation"
594 depends on DEBUG_KERNEL && !IA64
596 Enables the display of the minimum amount of free stack which each
597 task has ever had available in the sysrq-T and sysrq-P debug output.
599 This option will slow down process creation somewhat.
603 depends on DEBUG_KERNEL
605 Enable this to turn on extended checks in the virtual-memory system
606 that may impact performance.
610 config DEBUG_VM_VMACACHE
611 bool "Debug VMA caching"
614 Enable this to turn on VMA caching debug information. Doing so
615 can cause significant overhead, so only enable it in non-production
621 bool "Debug VM red-black trees"
624 Enable VM red-black tree debugging information and extra validations.
628 config DEBUG_VM_PGFLAGS
629 bool "Debug page-flags operations"
632 Enables extra validation on page flags operations.
636 config ARCH_HAS_DEBUG_VIRTUAL
640 bool "Debug VM translations"
641 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL
643 Enable some costly sanity checks in virtual to page code. This can
644 catch mistakes with virt_to_page() and friends.
648 config DEBUG_NOMMU_REGIONS
649 bool "Debug the global anon/private NOMMU mapping region tree"
650 depends on DEBUG_KERNEL && !MMU
652 This option causes the global tree of anonymous and private mapping
653 regions to be regularly checked for invalid topology.
655 config DEBUG_MEMORY_INIT
656 bool "Debug memory initialisation" if EXPERT
659 Enable this for additional checks during memory initialisation.
660 The sanity checks verify aspects of the VM such as the memory model
661 and other information provided by the architecture. Verbose
662 information will be printed at KERN_DEBUG loglevel depending
663 on the mminit_loglevel= command-line option.
667 config MEMORY_NOTIFIER_ERROR_INJECT
668 tristate "Memory hotplug notifier error injection module"
669 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION
671 This option provides the ability to inject artificial errors to
672 memory hotplug notifier chain callbacks. It is controlled through
673 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
675 If the notifier call chain should be failed with some events
676 notified, write the error code to "actions/<notifier event>/error".
678 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
680 # cd /sys/kernel/debug/notifier-error-inject/memory
681 # echo -12 > actions/MEM_GOING_OFFLINE/error
682 # echo offline > /sys/devices/system/memory/memoryXXX/state
683 bash: echo: write error: Cannot allocate memory
685 To compile this code as a module, choose M here: the module will
686 be called memory-notifier-error-inject.
690 config DEBUG_PER_CPU_MAPS
691 bool "Debug access to per_cpu maps"
692 depends on DEBUG_KERNEL
695 Say Y to verify that the per_cpu map being accessed has
696 been set up. This adds a fair amount of code to kernel memory
697 and decreases performance.
702 bool "Highmem debugging"
703 depends on DEBUG_KERNEL && HIGHMEM
705 This option enables additional error checking for high memory
706 systems. Disable for production systems.
708 config HAVE_DEBUG_STACKOVERFLOW
711 config DEBUG_STACKOVERFLOW
712 bool "Check for stack overflows"
713 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
715 Say Y here if you want to check for overflows of kernel, IRQ
716 and exception stacks (if your architecture uses them). This
717 option will show detailed messages if free stack space drops
718 below a certain limit.
720 These kinds of bugs usually occur when call-chains in the
721 kernel get too deep, especially when interrupts are
724 Use this in cases where you see apparently random memory
725 corruption, especially if it appears in 'struct thread_info'
727 If in doubt, say "N".
729 source "lib/Kconfig.kmemcheck"
731 source "lib/Kconfig.kasan"
733 endmenu # "Memory Debugging"
738 KCOV does not have any arch-specific code, but currently it is enabled
739 only for x86_64. KCOV requires testing on other archs, and most likely
740 disabling of instrumentation for some early boot code.
743 bool "Code coverage for fuzzing"
744 depends on ARCH_HAS_KCOV
746 select GCC_PLUGINS if !COMPILE_TEST
747 select GCC_PLUGIN_SANCOV if !COMPILE_TEST
749 KCOV exposes kernel code coverage information in a form suitable
750 for coverage-guided fuzzing (randomized testing).
752 If RANDOMIZE_BASE is enabled, PC values will not be stable across
753 different machines and across reboots. If you need stable PC values,
754 disable RANDOMIZE_BASE.
756 For more details, see Documentation/dev-tools/kcov.rst.
758 config KCOV_INSTRUMENT_ALL
759 bool "Instrument all code by default"
763 If you are doing generic system call fuzzing (like e.g. syzkaller),
764 then you will want to instrument the whole kernel and you should
765 say y here. If you are doing more targeted fuzzing (like e.g.
766 filesystem fuzzing with AFL) then you will want to enable coverage
767 for more specific subsets of files, and should say n here.
770 bool "Debug shared IRQ handlers"
771 depends on DEBUG_KERNEL
773 Enable this to generate a spurious interrupt as soon as a shared
774 interrupt handler is registered, and just before one is deregistered.
775 Drivers ought to be able to handle interrupts coming in at those
776 points; some don't and need to be caught.
778 menu "Debug Lockups and Hangs"
780 config LOCKUP_DETECTOR
781 bool "Detect Hard and Soft Lockups"
782 depends on DEBUG_KERNEL && !S390
784 Say Y here to enable the kernel to act as a watchdog to detect
785 hard and soft lockups.
787 Softlockups are bugs that cause the kernel to loop in kernel
788 mode for more than 20 seconds, without giving other tasks a
789 chance to run. The current stack trace is displayed upon
790 detection and the system will stay locked up.
792 Hardlockups are bugs that cause the CPU to loop in kernel mode
793 for more than 10 seconds, without letting other interrupts have a
794 chance to run. The current stack trace is displayed upon detection
795 and the system will stay locked up.
797 The overhead should be minimal. A periodic hrtimer runs to
798 generate interrupts and kick the watchdog task every 4 seconds.
799 An NMI is generated every 10 seconds or so to check for hardlockups.
801 The frequency of hrtimer and NMI events and the soft and hard lockup
802 thresholds can be controlled through the sysctl watchdog_thresh.
804 config HARDLOCKUP_DETECTOR
806 depends on LOCKUP_DETECTOR && !HAVE_NMI_WATCHDOG
807 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI
809 config BOOTPARAM_HARDLOCKUP_PANIC
810 bool "Panic (Reboot) On Hard Lockups"
811 depends on HARDLOCKUP_DETECTOR
813 Say Y here to enable the kernel to panic on "hard lockups",
814 which are bugs that cause the kernel to loop in kernel
815 mode with interrupts disabled for more than 10 seconds (configurable
816 using the watchdog_thresh sysctl).
820 config BOOTPARAM_HARDLOCKUP_PANIC_VALUE
822 depends on HARDLOCKUP_DETECTOR
824 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC
825 default 1 if BOOTPARAM_HARDLOCKUP_PANIC
827 config BOOTPARAM_SOFTLOCKUP_PANIC
828 bool "Panic (Reboot) On Soft Lockups"
829 depends on LOCKUP_DETECTOR
831 Say Y here to enable the kernel to panic on "soft lockups",
832 which are bugs that cause the kernel to loop in kernel
833 mode for more than 20 seconds (configurable using the watchdog_thresh
834 sysctl), without giving other tasks a chance to run.
836 The panic can be used in combination with panic_timeout,
837 to cause the system to reboot automatically after a
838 lockup has been detected. This feature is useful for
839 high-availability systems that have uptime guarantees and
840 where a lockup must be resolved ASAP.
844 config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
846 depends on LOCKUP_DETECTOR
848 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
849 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
851 config DETECT_HUNG_TASK
852 bool "Detect Hung Tasks"
853 depends on DEBUG_KERNEL
854 default LOCKUP_DETECTOR
856 Say Y here to enable the kernel to detect "hung tasks",
857 which are bugs that cause the task to be stuck in
858 uninterruptible "D" state indefinitely.
860 When a hung task is detected, the kernel will print the
861 current stack trace (which you should report), but the
862 task will stay in uninterruptible state. If lockdep is
863 enabled then all held locks will also be reported. This
864 feature has negligible overhead.
866 config DEFAULT_HUNG_TASK_TIMEOUT
867 int "Default timeout for hung task detection (in seconds)"
868 depends on DETECT_HUNG_TASK
871 This option controls the default timeout (in seconds) used
872 to determine when a task has become non-responsive and should
875 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
876 sysctl or by writing a value to
877 /proc/sys/kernel/hung_task_timeout_secs.
879 A timeout of 0 disables the check. The default is two minutes.
880 Keeping the default should be fine in most cases.
882 config BOOTPARAM_HUNG_TASK_PANIC
883 bool "Panic (Reboot) On Hung Tasks"
884 depends on DETECT_HUNG_TASK
886 Say Y here to enable the kernel to panic on "hung tasks",
887 which are bugs that cause the kernel to leave a task stuck
888 in uninterruptible "D" state.
890 The panic can be used in combination with panic_timeout,
891 to cause the system to reboot automatically after a
892 hung task has been detected. This feature is useful for
893 high-availability systems that have uptime guarantees and
894 where a hung tasks must be resolved ASAP.
898 config BOOTPARAM_HUNG_TASK_PANIC_VALUE
900 depends on DETECT_HUNG_TASK
902 default 0 if !BOOTPARAM_HUNG_TASK_PANIC
903 default 1 if BOOTPARAM_HUNG_TASK_PANIC
906 bool "Detect Workqueue Stalls"
907 depends on DEBUG_KERNEL
909 Say Y here to enable stall detection on workqueues. If a
910 worker pool doesn't make forward progress on a pending work
911 item for over a given amount of time, 30s by default, a
912 warning message is printed along with dump of workqueue
913 state. This can be configured through kernel parameter
914 "workqueue.watchdog_thresh" and its sysfs counterpart.
916 endmenu # "Debug lockups and hangs"
921 Say Y here to enable the kernel to panic when it oopses. This
922 has the same effect as setting oops=panic on the kernel command
925 This feature is useful to ensure that the kernel does not do
926 anything erroneous after an oops which could result in data
927 corruption or other issues.
931 config PANIC_ON_OOPS_VALUE
934 default 0 if !PANIC_ON_OOPS
935 default 1 if PANIC_ON_OOPS
941 Set the timeout value (in seconds) until a reboot occurs when the
942 the kernel panics. If n = 0, then we wait forever. A timeout
943 value n > 0 will wait n seconds before rebooting, while a timeout
944 value n < 0 will reboot immediately.
947 bool "Collect scheduler debugging info"
948 depends on DEBUG_KERNEL && PROC_FS
951 If you say Y here, the /proc/sched_debug file will be provided
952 that can help debug the scheduler. The runtime overhead of this
960 bool "Collect scheduler statistics"
961 depends on DEBUG_KERNEL && PROC_FS
964 If you say Y here, additional code will be inserted into the
965 scheduler and related routines to collect statistics about
966 scheduler behavior and provide them in /proc/schedstat. These
967 stats may be useful for both tuning and debugging the scheduler
968 If you aren't debugging the scheduler or trying to tune a specific
969 application, you can say N to avoid the very slight overhead
972 config SCHED_STACK_END_CHECK
973 bool "Detect stack corruption on calls to schedule()"
974 depends on DEBUG_KERNEL
977 This option checks for a stack overrun on calls to schedule().
978 If the stack end location is found to be over written always panic as
979 the content of the corrupted region can no longer be trusted.
980 This is to ensure no erroneous behaviour occurs which could result in
981 data corruption or a sporadic crash at a later stage once the region
982 is examined. The runtime overhead introduced is minimal.
984 config DEBUG_TIMEKEEPING
985 bool "Enable extra timekeeping sanity checking"
987 This option will enable additional timekeeping sanity checks
988 which may be helpful when diagnosing issues where timekeeping
989 problems are suspected.
991 This may include checks in the timekeeping hotpaths, so this
992 option may have a (very small) performance impact to some
998 bool "Debug preemptible kernel"
999 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT
1002 If you say Y here then the kernel will use a debug variant of the
1003 commonly used smp_processor_id() function and will print warnings
1004 if kernel code uses it in a preemption-unsafe way. Also, the kernel
1005 will detect preemption count underflows.
1007 menu "Lock Debugging (spinlocks, mutexes, etc...)"
1009 config DEBUG_RT_MUTEXES
1010 bool "RT Mutex debugging, deadlock detection"
1011 depends on DEBUG_KERNEL && RT_MUTEXES
1013 This allows rt mutex semantics violations and rt mutex related
1014 deadlocks (lockups) to be detected and reported automatically.
1016 config DEBUG_SPINLOCK
1017 bool "Spinlock and rw-lock debugging: basic checks"
1018 depends on DEBUG_KERNEL
1019 select UNINLINE_SPIN_UNLOCK
1021 Say Y here and build SMP to catch missing spinlock initialization
1022 and certain other kinds of spinlock errors commonly made. This is
1023 best used in conjunction with the NMI watchdog so that spinlock
1024 deadlocks are also debuggable.
1026 config DEBUG_MUTEXES
1027 bool "Mutex debugging: basic checks"
1028 depends on DEBUG_KERNEL
1030 This feature allows mutex semantics violations to be detected and
1033 config DEBUG_WW_MUTEX_SLOWPATH
1034 bool "Wait/wound mutex debugging: Slowpath testing"
1035 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1036 select DEBUG_LOCK_ALLOC
1037 select DEBUG_SPINLOCK
1038 select DEBUG_MUTEXES
1040 This feature enables slowpath testing for w/w mutex users by
1041 injecting additional -EDEADLK wound/backoff cases. Together with
1042 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1043 will test all possible w/w mutex interface abuse with the
1044 exception of simply not acquiring all the required locks.
1045 Note that this feature can introduce significant overhead, so
1046 it really should not be enabled in a production or distro kernel,
1047 even a debug kernel. If you are a driver writer, enable it. If
1048 you are a distro, do not.
1050 config DEBUG_LOCK_ALLOC
1051 bool "Lock debugging: detect incorrect freeing of live locks"
1052 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1053 select DEBUG_SPINLOCK
1054 select DEBUG_MUTEXES
1057 This feature will check whether any held lock (spinlock, rwlock,
1058 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1059 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1060 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1061 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1062 held during task exit.
1064 config PROVE_LOCKING
1065 bool "Lock debugging: prove locking correctness"
1066 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1068 select DEBUG_SPINLOCK
1069 select DEBUG_MUTEXES
1070 select DEBUG_LOCK_ALLOC
1071 select TRACE_IRQFLAGS
1074 This feature enables the kernel to prove that all locking
1075 that occurs in the kernel runtime is mathematically
1076 correct: that under no circumstance could an arbitrary (and
1077 not yet triggered) combination of observed locking
1078 sequences (on an arbitrary number of CPUs, running an
1079 arbitrary number of tasks and interrupt contexts) cause a
1082 In short, this feature enables the kernel to report locking
1083 related deadlocks before they actually occur.
1085 The proof does not depend on how hard and complex a
1086 deadlock scenario would be to trigger: how many
1087 participant CPUs, tasks and irq-contexts would be needed
1088 for it to trigger. The proof also does not depend on
1089 timing: if a race and a resulting deadlock is possible
1090 theoretically (no matter how unlikely the race scenario
1091 is), it will be proven so and will immediately be
1092 reported by the kernel (once the event is observed that
1093 makes the deadlock theoretically possible).
1095 If a deadlock is impossible (i.e. the locking rules, as
1096 observed by the kernel, are mathematically correct), the
1097 kernel reports nothing.
1099 NOTE: this feature can also be enabled for rwlocks, mutexes
1100 and rwsems - in which case all dependencies between these
1101 different locking variants are observed and mapped too, and
1102 the proof of observed correctness is also maintained for an
1103 arbitrary combination of these separate locking variants.
1105 For more details, see Documentation/locking/lockdep-design.txt.
1109 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1111 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE && !ARC && !SCORE
1115 config LOCKDEP_SMALL
1119 bool "Lock usage statistics"
1120 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1122 select DEBUG_SPINLOCK
1123 select DEBUG_MUTEXES
1124 select DEBUG_LOCK_ALLOC
1127 This feature enables tracking lock contention points
1129 For more details, see Documentation/locking/lockstat.txt
1131 This also enables lock events required by "perf lock",
1133 If you want to use "perf lock", you also need to turn on
1134 CONFIG_EVENT_TRACING.
1136 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1137 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1139 config DEBUG_LOCKDEP
1140 bool "Lock dependency engine debugging"
1141 depends on DEBUG_KERNEL && LOCKDEP
1143 If you say Y here, the lock dependency engine will do
1144 additional runtime checks to debug itself, at the price
1145 of more runtime overhead.
1147 config DEBUG_ATOMIC_SLEEP
1148 bool "Sleep inside atomic section checking"
1149 select PREEMPT_COUNT
1150 depends on DEBUG_KERNEL
1152 If you say Y here, various routines which may sleep will become very
1153 noisy if they are called inside atomic sections: when a spinlock is
1154 held, inside an rcu read side critical section, inside preempt disabled
1155 sections, inside an interrupt, etc...
1157 config DEBUG_LOCKING_API_SELFTESTS
1158 bool "Locking API boot-time self-tests"
1159 depends on DEBUG_KERNEL
1161 Say Y here if you want the kernel to run a short self-test during
1162 bootup. The self-test checks whether common types of locking bugs
1163 are detected by debugging mechanisms or not. (if you disable
1164 lock debugging then those bugs wont be detected of course.)
1165 The following locking APIs are covered: spinlocks, rwlocks,
1168 config LOCK_TORTURE_TEST
1169 tristate "torture tests for locking"
1170 depends on DEBUG_KERNEL
1174 This option provides a kernel module that runs torture tests
1175 on kernel locking primitives. The kernel module may be built
1176 after the fact on the running kernel to be tested, if desired.
1178 Say Y here if you want kernel locking-primitive torture tests
1179 to be built into the kernel.
1180 Say M if you want these torture tests to build as a module.
1181 Say N if you are unsure.
1183 config WW_MUTEX_SELFTEST
1184 tristate "Wait/wound mutex selftests"
1186 This option provides a kernel module that runs tests on the
1187 on the struct ww_mutex locking API.
1189 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction
1190 with this test harness.
1192 Say M if you want these self tests to build as a module.
1193 Say N if you are unsure.
1195 endmenu # lock debugging
1197 config TRACE_IRQFLAGS
1200 Enables hooks to interrupt enabling and disabling for
1201 either tracing or lock debugging.
1204 bool "Stack backtrace support"
1205 depends on STACKTRACE_SUPPORT
1207 This option causes the kernel to create a /proc/pid/stack for
1208 every process, showing its current stack trace.
1209 It is also used by various kernel debugging features that require
1210 stack trace generation.
1212 config DEBUG_KOBJECT
1213 bool "kobject debugging"
1214 depends on DEBUG_KERNEL
1216 If you say Y here, some extra kobject debugging messages will be sent
1219 config DEBUG_KOBJECT_RELEASE
1220 bool "kobject release debugging"
1221 depends on DEBUG_OBJECTS_TIMERS
1223 kobjects are reference counted objects. This means that their
1224 last reference count put is not predictable, and the kobject can
1225 live on past the point at which a driver decides to drop it's
1226 initial reference to the kobject gained on allocation. An
1227 example of this would be a struct device which has just been
1230 However, some buggy drivers assume that after such an operation,
1231 the memory backing the kobject can be immediately freed. This
1232 goes completely against the principles of a refcounted object.
1234 If you say Y here, the kernel will delay the release of kobjects
1235 on the last reference count to improve the visibility of this
1236 kind of kobject release bug.
1238 config HAVE_DEBUG_BUGVERBOSE
1241 config DEBUG_BUGVERBOSE
1242 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT
1243 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE)
1246 Say Y here to make BUG() panics output the file name and line number
1247 of the BUG call as well as the EIP and oops trace. This aids
1248 debugging but costs about 70-100K of memory.
1251 bool "Debug linked list manipulation"
1252 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION
1254 Enable this to turn on extended checks in the linked-list
1259 config DEBUG_PI_LIST
1260 bool "Debug priority linked list manipulation"
1261 depends on DEBUG_KERNEL
1263 Enable this to turn on extended checks in the priority-ordered
1264 linked-list (plist) walking routines. This checks the entire
1265 list multiple times during each manipulation.
1270 bool "Debug SG table operations"
1271 depends on DEBUG_KERNEL
1273 Enable this to turn on checks on scatter-gather tables. This can
1274 help find problems with drivers that do not properly initialize
1279 config DEBUG_NOTIFIERS
1280 bool "Debug notifier call chains"
1281 depends on DEBUG_KERNEL
1283 Enable this to turn on sanity checking for notifier call chains.
1284 This is most useful for kernel developers to make sure that
1285 modules properly unregister themselves from notifier chains.
1286 This is a relatively cheap check but if you care about maximum
1289 config DEBUG_CREDENTIALS
1290 bool "Debug credential management"
1291 depends on DEBUG_KERNEL
1293 Enable this to turn on some debug checking for credential
1294 management. The additional code keeps track of the number of
1295 pointers from task_structs to any given cred struct, and checks to
1296 see that this number never exceeds the usage count of the cred
1299 Furthermore, if SELinux is enabled, this also checks that the
1300 security pointer in the cred struct is never seen to be invalid.
1304 menu "RCU Debugging"
1307 def_bool PROVE_LOCKING
1309 config PROVE_RCU_REPEATEDLY
1310 bool "RCU debugging: don't disable PROVE_RCU on first splat"
1311 depends on PROVE_RCU
1314 By itself, PROVE_RCU will disable checking upon issuing the
1315 first warning (or "splat"). This feature prevents such
1316 disabling, allowing multiple RCU-lockdep warnings to be printed
1319 Say Y to allow multiple RCU-lockdep warnings per boot.
1321 Say N if you are unsure.
1323 config SPARSE_RCU_POINTER
1324 bool "RCU debugging: sparse-based checks for pointer usage"
1327 This feature enables the __rcu sparse annotation for
1328 RCU-protected pointers. This annotation will cause sparse
1329 to flag any non-RCU used of annotated pointers. This can be
1330 helpful when debugging RCU usage. Please note that this feature
1331 is not intended to enforce code cleanliness; it is instead merely
1334 Say Y to make sparse flag questionable use of RCU-protected pointers
1336 Say N if you are unsure.
1342 config RCU_PERF_TEST
1343 tristate "performance tests for RCU"
1344 depends on DEBUG_KERNEL
1350 This option provides a kernel module that runs performance
1351 tests on the RCU infrastructure. The kernel module may be built
1352 after the fact on the running kernel to be tested, if desired.
1354 Say Y here if you want RCU performance tests to be built into
1356 Say M if you want the RCU performance tests to build as a module.
1357 Say N if you are unsure.
1359 config RCU_TORTURE_TEST
1360 tristate "torture tests for RCU"
1361 depends on DEBUG_KERNEL
1367 This option provides a kernel module that runs torture tests
1368 on the RCU infrastructure. The kernel module may be built
1369 after the fact on the running kernel to be tested, if desired.
1371 Say Y here if you want RCU torture tests to be built into
1373 Say M if you want the RCU torture tests to build as a module.
1374 Say N if you are unsure.
1376 config RCU_TORTURE_TEST_SLOW_PREINIT
1377 bool "Slow down RCU grace-period pre-initialization to expose races"
1378 depends on RCU_TORTURE_TEST
1380 This option delays grace-period pre-initialization (the
1381 propagation of CPU-hotplug changes up the rcu_node combining
1382 tree) for a few jiffies between initializing each pair of
1383 consecutive rcu_node structures. This helps to expose races
1384 involving grace-period pre-initialization, in other words, it
1385 makes your kernel less stable. It can also greatly increase
1386 grace-period latency, especially on systems with large numbers
1387 of CPUs. This is useful when torture-testing RCU, but in
1388 almost no other circumstance.
1390 Say Y here if you want your system to crash and hang more often.
1391 Say N if you want a sane system.
1393 config RCU_TORTURE_TEST_SLOW_PREINIT_DELAY
1394 int "How much to slow down RCU grace-period pre-initialization"
1397 depends on RCU_TORTURE_TEST_SLOW_PREINIT
1399 This option specifies the number of jiffies to wait between
1400 each rcu_node structure pre-initialization step.
1402 config RCU_TORTURE_TEST_SLOW_INIT
1403 bool "Slow down RCU grace-period initialization to expose races"
1404 depends on RCU_TORTURE_TEST
1406 This option delays grace-period initialization for a few
1407 jiffies between initializing each pair of consecutive
1408 rcu_node structures. This helps to expose races involving
1409 grace-period initialization, in other words, it makes your
1410 kernel less stable. It can also greatly increase grace-period
1411 latency, especially on systems with large numbers of CPUs.
1412 This is useful when torture-testing RCU, but in almost no
1415 Say Y here if you want your system to crash and hang more often.
1416 Say N if you want a sane system.
1418 config RCU_TORTURE_TEST_SLOW_INIT_DELAY
1419 int "How much to slow down RCU grace-period initialization"
1422 depends on RCU_TORTURE_TEST_SLOW_INIT
1424 This option specifies the number of jiffies to wait between
1425 each rcu_node structure initialization.
1427 config RCU_TORTURE_TEST_SLOW_CLEANUP
1428 bool "Slow down RCU grace-period cleanup to expose races"
1429 depends on RCU_TORTURE_TEST
1431 This option delays grace-period cleanup for a few jiffies
1432 between cleaning up each pair of consecutive rcu_node
1433 structures. This helps to expose races involving grace-period
1434 cleanup, in other words, it makes your kernel less stable.
1435 It can also greatly increase grace-period latency, especially
1436 on systems with large numbers of CPUs. This is useful when
1437 torture-testing RCU, but in almost no other circumstance.
1439 Say Y here if you want your system to crash and hang more often.
1440 Say N if you want a sane system.
1442 config RCU_TORTURE_TEST_SLOW_CLEANUP_DELAY
1443 int "How much to slow down RCU grace-period cleanup"
1446 depends on RCU_TORTURE_TEST_SLOW_CLEANUP
1448 This option specifies the number of jiffies to wait between
1449 each rcu_node structure cleanup operation.
1451 config RCU_CPU_STALL_TIMEOUT
1452 int "RCU CPU stall timeout in seconds"
1453 depends on RCU_STALL_COMMON
1457 If a given RCU grace period extends more than the specified
1458 number of seconds, a CPU stall warning is printed. If the
1459 RCU grace period persists, additional CPU stall warnings are
1460 printed at more widely spaced intervals.
1463 bool "Enable tracing for RCU"
1464 depends on DEBUG_KERNEL
1465 default y if TREE_RCU
1468 This option provides tracing in RCU which presents stats
1469 in debugfs for debugging RCU implementation. It also enables
1470 additional tracepoints for ftrace-style event tracing.
1472 Say Y here if you want to enable RCU tracing
1473 Say N if you are unsure.
1475 config RCU_EQS_DEBUG
1476 bool "Provide debugging asserts for adding NO_HZ support to an arch"
1477 depends on DEBUG_KERNEL
1479 This option provides consistency checks in RCU's handling of
1480 NO_HZ. These checks have proven quite helpful in detecting
1481 bugs in arch-specific NO_HZ code.
1483 Say N here if you need ultimate kernel/user switch latencies
1484 Say Y if you are unsure
1486 endmenu # "RCU Debugging"
1488 config DEBUG_WQ_FORCE_RR_CPU
1489 bool "Force round-robin CPU selection for unbound work items"
1490 depends on DEBUG_KERNEL
1493 Workqueue used to implicitly guarantee that work items queued
1494 without explicit CPU specified are put on the local CPU. This
1495 guarantee is no longer true and while local CPU is still
1496 preferred work items may be put on foreign CPUs. Kernel
1497 parameter "workqueue.debug_force_rr_cpu" is added to force
1498 round-robin CPU selection to flush out usages which depend on the
1499 now broken guarantee. This config option enables the debug
1500 feature by default. When enabled, memory and cache locality will
1503 config DEBUG_BLOCK_EXT_DEVT
1504 bool "Force extended block device numbers and spread them"
1505 depends on DEBUG_KERNEL
1509 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON
1510 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT
1511 YOU ARE DOING. Distros, please enable this and fix whatever
1514 Conventionally, block device numbers are allocated from
1515 predetermined contiguous area. However, extended block area
1516 may introduce non-contiguous block device numbers. This
1517 option forces most block device numbers to be allocated from
1518 the extended space and spreads them to discover kernel or
1519 userland code paths which assume predetermined contiguous
1520 device number allocation.
1522 Note that turning on this debug option shuffles all the
1523 device numbers for all IDE and SCSI devices including libata
1524 ones, so root partition specified using device number
1525 directly (via rdev or root=MAJ:MIN) won't work anymore.
1526 Textual device names (root=/dev/sdXn) will continue to work.
1528 Say N if you are unsure.
1530 config CPU_HOTPLUG_STATE_CONTROL
1531 bool "Enable CPU hotplug state control"
1532 depends on DEBUG_KERNEL
1533 depends on HOTPLUG_CPU
1536 Allows to write steps between "offline" and "online" to the CPUs
1537 sysfs target file so states can be stepped granular. This is a debug
1538 option for now as the hotplug machinery cannot be stopped and
1539 restarted at arbitrary points yet.
1541 Say N if your are unsure.
1543 config NOTIFIER_ERROR_INJECTION
1544 tristate "Notifier error injection"
1545 depends on DEBUG_KERNEL
1548 This option provides the ability to inject artificial errors to
1549 specified notifier chain callbacks. It is useful to test the error
1550 handling of notifier call chain failures.
1554 config PM_NOTIFIER_ERROR_INJECT
1555 tristate "PM notifier error injection module"
1556 depends on PM && NOTIFIER_ERROR_INJECTION
1557 default m if PM_DEBUG
1559 This option provides the ability to inject artificial errors to
1560 PM notifier chain callbacks. It is controlled through debugfs
1561 interface /sys/kernel/debug/notifier-error-inject/pm
1563 If the notifier call chain should be failed with some events
1564 notified, write the error code to "actions/<notifier event>/error".
1566 Example: Inject PM suspend error (-12 = -ENOMEM)
1568 # cd /sys/kernel/debug/notifier-error-inject/pm/
1569 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1570 # echo mem > /sys/power/state
1571 bash: echo: write error: Cannot allocate memory
1573 To compile this code as a module, choose M here: the module will
1574 be called pm-notifier-error-inject.
1578 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1579 tristate "OF reconfig notifier error injection module"
1580 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1582 This option provides the ability to inject artificial errors to
1583 OF reconfig notifier chain callbacks. It is controlled
1584 through debugfs interface under
1585 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1587 If the notifier call chain should be failed with some events
1588 notified, write the error code to "actions/<notifier event>/error".
1590 To compile this code as a module, choose M here: the module will
1591 be called of-reconfig-notifier-error-inject.
1595 config NETDEV_NOTIFIER_ERROR_INJECT
1596 tristate "Netdev notifier error injection module"
1597 depends on NET && NOTIFIER_ERROR_INJECTION
1599 This option provides the ability to inject artificial errors to
1600 netdevice notifier chain callbacks. It is controlled through debugfs
1601 interface /sys/kernel/debug/notifier-error-inject/netdev
1603 If the notifier call chain should be failed with some events
1604 notified, write the error code to "actions/<notifier event>/error".
1606 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1608 # cd /sys/kernel/debug/notifier-error-inject/netdev
1609 # echo -22 > actions/NETDEV_CHANGEMTU/error
1610 # ip link set eth0 mtu 1024
1611 RTNETLINK answers: Invalid argument
1613 To compile this code as a module, choose M here: the module will
1614 be called netdev-notifier-error-inject.
1618 config FAULT_INJECTION
1619 bool "Fault-injection framework"
1620 depends on DEBUG_KERNEL
1622 Provide fault-injection framework.
1623 For more details, see Documentation/fault-injection/.
1626 bool "Fault-injection capability for kmalloc"
1627 depends on FAULT_INJECTION
1628 depends on SLAB || SLUB
1630 Provide fault-injection capability for kmalloc.
1632 config FAIL_PAGE_ALLOC
1633 bool "Fault-injection capabilitiy for alloc_pages()"
1634 depends on FAULT_INJECTION
1636 Provide fault-injection capability for alloc_pages().
1638 config FAIL_MAKE_REQUEST
1639 bool "Fault-injection capability for disk IO"
1640 depends on FAULT_INJECTION && BLOCK
1642 Provide fault-injection capability for disk IO.
1644 config FAIL_IO_TIMEOUT
1645 bool "Fault-injection capability for faking disk interrupts"
1646 depends on FAULT_INJECTION && BLOCK
1648 Provide fault-injection capability on end IO handling. This
1649 will make the block layer "forget" an interrupt as configured,
1650 thus exercising the error handling.
1652 Only works with drivers that use the generic timeout handling,
1653 for others it wont do anything.
1655 config FAIL_MMC_REQUEST
1656 bool "Fault-injection capability for MMC IO"
1657 depends on FAULT_INJECTION_DEBUG_FS && MMC
1659 Provide fault-injection capability for MMC IO.
1660 This will make the mmc core return data errors. This is
1661 useful to test the error handling in the mmc block device
1662 and to test how the mmc host driver handles retries from
1666 bool "Fault-injection capability for futexes"
1668 depends on FAULT_INJECTION && FUTEX
1670 Provide fault-injection capability for futexes.
1672 config FAULT_INJECTION_DEBUG_FS
1673 bool "Debugfs entries for fault-injection capabilities"
1674 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1676 Enable configuration of fault-injection capabilities via debugfs.
1678 config FAULT_INJECTION_STACKTRACE_FILTER
1679 bool "stacktrace filter for fault-injection capabilities"
1680 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1683 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !SCORE
1685 Provide stacktrace filter for fault-injection capabilities
1688 bool "Latency measuring infrastructure"
1689 depends on DEBUG_KERNEL
1690 depends on STACKTRACE_SUPPORT
1692 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC
1699 Enable this option if you want to use the LatencyTOP tool
1700 to find out which userspace is blocking on what kernel operations.
1702 source kernel/trace/Kconfig
1704 menu "Runtime Testing"
1707 tristate "Linux Kernel Dump Test Tool Module"
1712 This module enables testing of the different dumping mechanisms by
1713 inducing system failures at predefined crash points.
1714 If you don't need it: say N
1715 Choose M here to compile this code as a module. The module will be
1718 Documentation on how to use the module can be found in
1719 Documentation/fault-injection/provoke-crashes.txt
1721 config TEST_LIST_SORT
1722 tristate "Linked list sorting test"
1723 depends on DEBUG_KERNEL || m
1725 Enable this to turn on 'list_sort()' function test. This test is
1726 executed only once during system boot (so affects only boot time),
1727 or at module load time.
1732 tristate "Array-based sort test"
1733 depends on DEBUG_KERNEL || m
1735 This option enables the self-test function of 'sort()' at boot,
1736 or at module load time.
1740 config KPROBES_SANITY_TEST
1741 bool "Kprobes sanity tests"
1742 depends on DEBUG_KERNEL
1746 This option provides for testing basic kprobes functionality on
1747 boot. A sample kprobe, jprobe and kretprobe are inserted and
1748 verified for functionality.
1750 Say N if you are unsure.
1752 config BACKTRACE_SELF_TEST
1753 tristate "Self test for the backtrace code"
1754 depends on DEBUG_KERNEL
1757 This option provides a kernel module that can be used to test
1758 the kernel stack backtrace code. This option is not useful
1759 for distributions or general kernels, but only for kernel
1760 developers working on architecture code.
1762 Note that if you want to also test saved backtraces, you will
1763 have to enable STACKTRACE as well.
1765 Say N if you are unsure.
1768 tristate "Red-Black tree test"
1769 depends on DEBUG_KERNEL
1771 A benchmark measuring the performance of the rbtree library.
1772 Also includes rbtree invariant checks.
1774 config INTERVAL_TREE_TEST
1775 tristate "Interval tree test"
1776 depends on m && DEBUG_KERNEL
1777 select INTERVAL_TREE
1779 A benchmark measuring the performance of the interval tree library
1782 tristate "Per cpu operations test"
1783 depends on m && DEBUG_KERNEL
1785 Enable this option to build test module which validates per-cpu
1790 config ATOMIC64_SELFTEST
1791 tristate "Perform an atomic64_t self-test"
1793 Enable this option to test the atomic64_t functions at boot or
1794 at module load time.
1798 config ASYNC_RAID6_TEST
1799 tristate "Self test for hardware accelerated raid6 recovery"
1800 depends on ASYNC_RAID6_RECOV
1803 This is a one-shot self test that permutes through the
1804 recovery of all the possible two disk failure scenarios for a
1805 N-disk array. Recovery is performed with the asynchronous
1806 raid6 recovery routines, and will optionally use an offload
1807 engine if one is available.
1812 tristate "Test functions located in the hexdump module at runtime"
1814 config TEST_STRING_HELPERS
1815 tristate "Test functions located in the string_helpers module at runtime"
1818 tristate "Test kstrto*() family of functions at runtime"
1821 tristate "Test printf() family of functions at runtime"
1824 tristate "Test bitmap_*() family of functions at runtime"
1827 Enable this option to test the bitmap functions at boot.
1832 tristate "Test functions located in the uuid module at runtime"
1834 config TEST_RHASHTABLE
1835 tristate "Perform selftest on resizable hash table"
1838 Enable this option to test the rhashtable functions at boot.
1843 tristate "Perform selftest on hash functions"
1846 Enable this option to test the kernel's integer (<linux/hash.h>),
1847 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>)
1848 hash functions on boot (or module load).
1850 This is intended to help people writing architecture-specific
1851 optimized versions. If unsure, say N.
1854 tristate "Perform selftest on priority array manager"
1858 Enable this option to test priority array manager on boot
1863 endmenu # runtime tests
1865 config PROVIDE_OHCI1394_DMA_INIT
1866 bool "Remote debugging over FireWire early on boot"
1867 depends on PCI && X86
1869 If you want to debug problems which hang or crash the kernel early
1870 on boot and the crashing machine has a FireWire port, you can use
1871 this feature to remotely access the memory of the crashed machine
1872 over FireWire. This employs remote DMA as part of the OHCI1394
1873 specification which is now the standard for FireWire controllers.
1875 With remote DMA, you can monitor the printk buffer remotely using
1876 firescope and access all memory below 4GB using fireproxy from gdb.
1877 Even controlling a kernel debugger is possible using remote DMA.
1881 If ohci1394_dma=early is used as boot parameter, it will initialize
1882 all OHCI1394 controllers which are found in the PCI config space.
1884 As all changes to the FireWire bus such as enabling and disabling
1885 devices cause a bus reset and thereby disable remote DMA for all
1886 devices, be sure to have the cable plugged and FireWire enabled on
1887 the debugging host before booting the debug target for debugging.
1889 This code (~1k) is freed after boot. By then, the firewire stack
1890 in charge of the OHCI-1394 controllers should be used instead.
1892 See Documentation/debugging-via-ohci1394.txt for more information.
1894 config DMA_API_DEBUG
1895 bool "Enable debugging of DMA-API usage"
1896 depends on HAVE_DMA_API_DEBUG
1898 Enable this option to debug the use of the DMA API by device drivers.
1899 With this option you will be able to detect common bugs in device
1900 drivers like double-freeing of DMA mappings or freeing mappings that
1901 were never allocated.
1903 This also attempts to catch cases where a page owned by DMA is
1904 accessed by the cpu in a way that could cause data corruption. For
1905 example, this enables cow_user_page() to check that the source page is
1908 This option causes a performance degradation. Use only if you want to
1909 debug device drivers and dma interactions.
1914 tristate "Test module loading with 'hello world' module"
1918 This builds the "test_module" module that emits "Hello, world"
1919 on printk when loaded. It is designed to be used for basic
1920 evaluation of the module loading subsystem (for example when
1921 validating module verification). It lacks any extra dependencies,
1922 and will not normally be loaded by the system unless explicitly
1927 config TEST_USER_COPY
1928 tristate "Test user/kernel boundary protections"
1932 This builds the "test_user_copy" module that runs sanity checks
1933 on the copy_to/from_user infrastructure, making sure basic
1934 user/kernel boundary testing is working. If it fails to load,
1935 a regression has been detected in the user/kernel memory boundary
1941 tristate "Test BPF filter functionality"
1945 This builds the "test_bpf" module that runs various test vectors
1946 against the BPF interpreter or BPF JIT compiler depending on the
1947 current setting. This is in particular useful for BPF JIT compiler
1948 development, but also to run regression tests against changes in
1949 the interpreter code. It also enables test stubs for eBPF maps and
1950 verifier used by user space verifier testsuite.
1954 config TEST_FIRMWARE
1955 tristate "Test firmware loading via userspace interface"
1957 depends on FW_LOADER
1959 This builds the "test_firmware" module that creates a userspace
1960 interface for testing firmware loading. This can be used to
1961 control the triggering of firmware loading without needing an
1962 actual firmware-using device. The contents can be rechecked by
1968 tristate "udelay test driver"
1971 This builds the "udelay_test" module that helps to make sure
1972 that udelay() is working properly.
1978 depends on HAVE_MEMBLOCK
1980 This option adds a kernel parameter 'memtest', which allows memtest
1982 memtest=0, mean disabled; -- default
1983 memtest=1, mean do 1 test pattern;
1985 memtest=17, mean do 17 test patterns.
1986 If you are unsure how to answer this question, answer N.
1988 config TEST_STATIC_KEYS
1989 tristate "Test static keys"
1993 Test the static key interfaces.
1997 config BUG_ON_DATA_CORRUPTION
1998 bool "Trigger a BUG when data corruption is detected"
2001 Select this option if the kernel should BUG when it encounters
2002 data corruption in kernel memory structures when they get checked
2007 source "samples/Kconfig"
2009 source "lib/Kconfig.kgdb"
2011 source "lib/Kconfig.ubsan"
2013 config ARCH_HAS_DEVMEM_IS_ALLOWED
2016 config STRICT_DEVMEM
2017 bool "Filter access to /dev/mem"
2018 depends on MMU && DEVMEM
2019 depends on ARCH_HAS_DEVMEM_IS_ALLOWED
2020 default y if TILE || PPC
2022 If this option is disabled, you allow userspace (root) access to all
2023 of memory, including kernel and userspace memory. Accidental
2024 access to this is obviously disastrous, but specific access can
2025 be used by people debugging the kernel. Note that with PAT support
2026 enabled, even in this case there are restrictions on /dev/mem
2027 use due to the cache aliasing requirements.
2029 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
2030 file only allows userspace access to PCI space and the BIOS code and
2031 data regions. This is sufficient for dosemu and X and all common
2036 config IO_STRICT_DEVMEM
2037 bool "Filter I/O access to /dev/mem"
2038 depends on STRICT_DEVMEM
2040 If this option is disabled, you allow userspace (root) access to all
2041 io-memory regardless of whether a driver is actively using that
2042 range. Accidental access to this is obviously disastrous, but
2043 specific access can be used by people debugging kernel drivers.
2045 If this option is switched on, the /dev/mem file only allows
2046 userspace access to *idle* io-memory ranges (see /proc/iomem) This
2047 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
2048 if the driver using a given range cannot be disabled.