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/kernel-parameters.txt
18 config MESSAGE_LOGLEVEL_DEFAULT
19 int "Default message log level (1-7)"
23 Default log level for printk statements with no specified priority.
25 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks
26 that are auditing their logs closely may want to set it to a lower
29 config BOOT_PRINTK_DELAY
30 bool "Delay each boot printk message by N milliseconds"
31 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY
33 This build option allows you to read kernel boot messages
34 by inserting a short delay after each one. The delay is
35 specified in milliseconds on the kernel command line,
38 It is likely that you would also need to use "lpj=M" to preset
39 the "loops per jiffie" value.
40 See a previous boot log for the "lpj" value to use for your
41 system, and then set "lpj=M" before setting "boot_delay=N".
42 NOTE: Using this option may adversely affect SMP systems.
43 I.e., processors other than the first one may not boot up.
44 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect
45 what it believes to be lockup conditions.
48 bool "Enable dynamic printk() support"
54 Compiles debug level messages into the kernel, which would not
55 otherwise be available at runtime. These messages can then be
56 enabled/disabled based on various levels of scope - per source file,
57 function, module, format string, and line number. This mechanism
58 implicitly compiles in all pr_debug() and dev_dbg() calls, which
59 enlarges the kernel text size by about 2%.
61 If a source file is compiled with DEBUG flag set, any
62 pr_debug() calls in it are enabled by default, but can be
63 disabled at runtime as below. Note that DEBUG flag is
64 turned on by many CONFIG_*DEBUG* options.
68 Dynamic debugging is controlled via the 'dynamic_debug/control' file,
69 which is contained in the 'debugfs' filesystem. Thus, the debugfs
70 filesystem must first be mounted before making use of this feature.
71 We refer the control file as: <debugfs>/dynamic_debug/control. This
72 file contains a list of the debug statements that can be enabled. The
73 format for each line of the file is:
75 filename:lineno [module]function flags format
77 filename : source file of the debug statement
78 lineno : line number of the debug statement
79 module : module that contains the debug statement
80 function : function that contains the debug statement
81 flags : '=p' means the line is turned 'on' for printing
82 format : the format used for the debug statement
86 nullarbor:~ # cat <debugfs>/dynamic_debug/control
87 # filename:lineno [module]function flags format
88 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012"
89 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012"
90 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012"
94 // enable the message at line 1603 of file svcsock.c
95 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' >
96 <debugfs>/dynamic_debug/control
98 // enable all the messages in file svcsock.c
99 nullarbor:~ # echo -n 'file svcsock.c +p' >
100 <debugfs>/dynamic_debug/control
102 // enable all the messages in the NFS server module
103 nullarbor:~ # echo -n 'module nfsd +p' >
104 <debugfs>/dynamic_debug/control
106 // enable all 12 messages in the function svc_process()
107 nullarbor:~ # echo -n 'func svc_process +p' >
108 <debugfs>/dynamic_debug/control
110 // disable all 12 messages in the function svc_process()
111 nullarbor:~ # echo -n 'func svc_process -p' >
112 <debugfs>/dynamic_debug/control
114 See Documentation/dynamic-debug-howto.txt for additional information.
116 endmenu # "printk and dmesg options"
118 menu "Compile-time checks and compiler options"
121 bool "Compile the kernel with debug info"
122 depends on DEBUG_KERNEL && !COMPILE_TEST
124 If you say Y here the resulting kernel image will include
125 debugging info resulting in a larger kernel image.
126 This adds debug symbols to the kernel and modules (gcc -g), and
127 is needed if you intend to use kernel crashdump or binary object
128 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
129 Say Y here only if you plan to debug the kernel.
133 config DEBUG_INFO_REDUCED
134 bool "Reduce debugging information"
135 depends on DEBUG_INFO
137 If you say Y here gcc is instructed to generate less debugging
138 information for structure types. This means that tools that
139 need full debugging information (like kgdb or systemtap) won't
140 be happy. But if you merely need debugging information to
141 resolve line numbers there is no loss. Advantage is that
142 build directory object sizes shrink dramatically over a full
143 DEBUG_INFO build and compile times are reduced too.
144 Only works with newer gcc versions.
146 config DEBUG_INFO_SPLIT
147 bool "Produce split debuginfo in .dwo files"
148 depends on DEBUG_INFO
150 Generate debug info into separate .dwo files. This significantly
151 reduces the build directory size for builds with DEBUG_INFO,
152 because it stores the information only once on disk in .dwo
153 files instead of multiple times in object files and executables.
154 In addition the debug information is also compressed.
156 Requires recent gcc (4.7+) and recent gdb/binutils.
157 Any tool that packages or reads debug information would need
158 to know about the .dwo files and include them.
159 Incompatible with older versions of ccache.
161 config DEBUG_INFO_DWARF4
162 bool "Generate dwarf4 debuginfo"
163 depends on DEBUG_INFO
165 Generate dwarf4 debug info. This requires recent versions
166 of gcc and gdb. It makes the debug information larger.
167 But it significantly improves the success of resolving
168 variables in gdb on optimized code.
171 bool "Provide GDB scripts for kernel debugging"
172 depends on DEBUG_INFO
174 This creates the required links to GDB helper scripts in the
175 build directory. If you load vmlinux into gdb, the helper
176 scripts will be automatically imported by gdb as well, and
177 additional functions are available to analyze a Linux kernel
178 instance. See Documentation/gdb-kernel-debugging.txt for further
181 config ENABLE_WARN_DEPRECATED
182 bool "Enable __deprecated logic"
185 Enable the __deprecated logic in the kernel build.
186 Disable this to suppress the "warning: 'foo' is deprecated
187 (declared at kernel/power/somefile.c:1234)" messages.
189 config ENABLE_MUST_CHECK
190 bool "Enable __must_check logic"
193 Enable the __must_check logic in the kernel build. Disable this to
194 suppress the "warning: ignoring return value of 'foo', declared with
195 attribute warn_unused_result" messages.
198 int "Warn for stack frames larger than (needs gcc 4.4)"
201 default 1024 if !64BIT
202 default 2048 if 64BIT
204 Tell gcc to warn at build time for stack frames larger than this.
205 Setting this too low will cause a lot of warnings.
206 Setting it to 0 disables the warning.
209 config STRIP_ASM_SYMS
210 bool "Strip assembler-generated symbols during link"
213 Strip internal assembler-generated symbols during a link (symbols
214 that look like '.Lxxx') so they don't pollute the output of
215 get_wchan() and suchlike.
218 bool "Generate readable assembler code"
219 depends on DEBUG_KERNEL
221 Disable some compiler optimizations that tend to generate human unreadable
222 assembler output. This may make the kernel slightly slower, but it helps
223 to keep kernel developers who have to stare a lot at assembler listings
226 config UNUSED_SYMBOLS
227 bool "Enable unused/obsolete exported symbols"
230 Unused but exported symbols make the kernel needlessly bigger. For
231 that reason most of these unused exports will soon be removed. This
232 option is provided temporarily to provide a transition period in case
233 some external kernel module needs one of these symbols anyway. If you
234 encounter such a case in your module, consider if you are actually
235 using the right API. (rationale: since nobody in the kernel is using
236 this in a module, there is a pretty good chance it's actually the
237 wrong interface to use). If you really need the symbol, please send a
238 mail to the linux kernel mailing list mentioning the symbol and why
239 you really need it, and what the merge plan to the mainline kernel for
243 bool "Track page owner"
244 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
247 select PAGE_EXTENSION
249 This keeps track of what call chain is the owner of a page, may
250 help to find bare alloc_page(s) leaks. Even if you include this
251 feature on your build, it is disabled in default. You should pass
252 "page_owner=on" to boot parameter in order to enable it. Eats
253 a fair amount of memory if enabled. See tools/vm/page_owner_sort.c
254 for user-space helper.
259 bool "Debug Filesystem"
261 debugfs is a virtual file system that kernel developers use to put
262 debugging files into. Enable this option to be able to read and
263 write to these files.
265 For detailed documentation on the debugfs API, see
266 Documentation/DocBook/filesystems.
271 bool "Run 'make headers_check' when building vmlinux"
274 This option will extract the user-visible kernel headers whenever
275 building the kernel, and will run basic sanity checks on them to
276 ensure that exported files do not attempt to include files which
277 were not exported, etc.
279 If you're making modifications to header files which are
280 relevant for userspace, say 'Y', and check the headers
281 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in
282 your build tree), to make sure they're suitable.
284 config DEBUG_SECTION_MISMATCH
285 bool "Enable full Section mismatch analysis"
287 The section mismatch analysis checks if there are illegal
288 references from one section to another section.
289 During linktime or runtime, some sections are dropped;
290 any use of code/data previously in these sections would
291 most likely result in an oops.
292 In the code, functions and variables are annotated with
293 __init,, etc. (see the full list in include/linux/init.h),
294 which results in the code/data being placed in specific sections.
295 The section mismatch analysis is always performed after a full
296 kernel build, and enabling this option causes the following
297 additional steps to occur:
298 - Add the option -fno-inline-functions-called-once to gcc commands.
299 When inlining a function annotated with __init in a non-init
300 function, we would lose the section information and thus
301 the analysis would not catch the illegal reference.
302 This option tells gcc to inline less (but it does result in
304 - Run the section mismatch analysis for each module/built-in.o file.
305 When we run the section mismatch analysis on vmlinux.o, we
306 lose valueble information about where the mismatch was
308 Running the analysis for each module/built-in.o file
309 tells where the mismatch happens much closer to the
310 source. The drawback is that the same mismatch is
311 reported at least twice.
312 - Enable verbose reporting from modpost in order to help resolve
313 the section mismatches that are reported.
315 config SECTION_MISMATCH_WARN_ONLY
316 bool "Make section mismatch errors non-fatal"
319 If you say N here, the build process will fail if there are any
320 section mismatch, instead of just throwing warnings.
325 # Select this config option from the architecture Kconfig, if it
326 # is preferred to always offer frame pointers as a config
327 # option on the architecture (regardless of KERNEL_DEBUG):
329 config ARCH_WANT_FRAME_POINTERS
334 bool "Compile the kernel with frame pointers"
335 depends on DEBUG_KERNEL && \
336 (CRIS || M68K || FRV || UML || \
337 AVR32 || SUPERH || BLACKFIN || MN10300 || METAG) || \
338 ARCH_WANT_FRAME_POINTERS
339 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
341 If you say Y here the resulting kernel image will be slightly
342 larger and slower, but it gives very useful debugging information
343 in case of kernel bugs. (precise oopses/stacktraces/warnings)
345 config STACK_VALIDATION
346 bool "Compile-time stack metadata validation"
347 depends on HAVE_STACK_VALIDATION
350 Add compile-time checks to validate stack metadata, including frame
351 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure
352 that runtime stack traces are more reliable.
354 For more information, see
355 tools/objtool/Documentation/stack-validation.txt.
357 config DEBUG_FORCE_WEAK_PER_CPU
358 bool "Force weak per-cpu definitions"
359 depends on DEBUG_KERNEL
361 s390 and alpha require percpu variables in modules to be
362 defined weak to work around addressing range issue which
363 puts the following two restrictions on percpu variable
366 1. percpu symbols must be unique whether static or not
367 2. percpu variables can't be defined inside a function
369 To ensure that generic code follows the above rules, this
370 option forces all percpu variables to be defined as weak.
372 endmenu # "Compiler options"
375 bool "Magic SysRq key"
378 If you say Y here, you will have some control over the system even
379 if the system crashes for example during kernel debugging (e.g., you
380 will be able to flush the buffer cache to disk, reboot the system
381 immediately or dump some status information). This is accomplished
382 by pressing various keys while holding SysRq (Alt+PrintScreen). It
383 also works on a serial console (on PC hardware at least), if you
384 send a BREAK and then within 5 seconds a command keypress. The
385 keys are documented in <file:Documentation/sysrq.txt>. Don't say Y
386 unless you really know what this hack does.
388 config MAGIC_SYSRQ_DEFAULT_ENABLE
389 hex "Enable magic SysRq key functions by default"
390 depends on MAGIC_SYSRQ
393 Specifies which SysRq key functions are enabled by default.
394 This may be set to 1 or 0 to enable or disable them all, or
395 to a bitmask as described in Documentation/sysrq.txt.
398 bool "Kernel debugging"
400 Say Y here if you are developing drivers or trying to debug and
401 identify kernel problems.
403 menu "Memory Debugging"
405 source mm/Kconfig.debug
408 bool "Debug object operations"
409 depends on DEBUG_KERNEL
411 If you say Y here, additional code will be inserted into the
412 kernel to track the life time of various objects and validate
413 the operations on those objects.
415 config DEBUG_OBJECTS_SELFTEST
416 bool "Debug objects selftest"
417 depends on DEBUG_OBJECTS
419 This enables the selftest of the object debug code.
421 config DEBUG_OBJECTS_FREE
422 bool "Debug objects in freed memory"
423 depends on DEBUG_OBJECTS
425 This enables checks whether a k/v free operation frees an area
426 which contains an object which has not been deactivated
427 properly. This can make kmalloc/kfree-intensive workloads
430 config DEBUG_OBJECTS_TIMERS
431 bool "Debug timer objects"
432 depends on DEBUG_OBJECTS
434 If you say Y here, additional code will be inserted into the
435 timer routines to track the life time of timer objects and
436 validate the timer operations.
438 config DEBUG_OBJECTS_WORK
439 bool "Debug work objects"
440 depends on DEBUG_OBJECTS
442 If you say Y here, additional code will be inserted into the
443 work queue routines to track the life time of work objects and
444 validate the work operations.
446 config DEBUG_OBJECTS_RCU_HEAD
447 bool "Debug RCU callbacks objects"
448 depends on DEBUG_OBJECTS
450 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
452 config DEBUG_OBJECTS_PERCPU_COUNTER
453 bool "Debug percpu counter objects"
454 depends on DEBUG_OBJECTS
456 If you say Y here, additional code will be inserted into the
457 percpu counter routines to track the life time of percpu counter
458 objects and validate the percpu counter operations.
460 config DEBUG_OBJECTS_ENABLE_DEFAULT
461 int "debug_objects bootup default value (0-1)"
464 depends on DEBUG_OBJECTS
466 Debug objects boot parameter default value
469 bool "Debug slab memory allocations"
470 depends on DEBUG_KERNEL && SLAB && !KMEMCHECK
472 Say Y here to have the kernel do limited verification on memory
473 allocation as well as poisoning memory on free to catch use of freed
474 memory. This can make kmalloc/kfree-intensive workloads much slower.
476 config DEBUG_SLAB_LEAK
477 bool "Memory leak debugging"
478 depends on DEBUG_SLAB
481 bool "SLUB debugging on by default"
482 depends on SLUB && SLUB_DEBUG && !KMEMCHECK
485 Boot with debugging on by default. SLUB boots by default with
486 the runtime debug capabilities switched off. Enabling this is
487 equivalent to specifying the "slub_debug" parameter on boot.
488 There is no support for more fine grained debug control like
489 possible with slub_debug=xxx. SLUB debugging may be switched
490 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
495 bool "Enable SLUB performance statistics"
496 depends on SLUB && SYSFS
498 SLUB statistics are useful to debug SLUBs allocation behavior in
499 order find ways to optimize the allocator. This should never be
500 enabled for production use since keeping statistics slows down
501 the allocator by a few percentage points. The slabinfo command
502 supports the determination of the most active slabs to figure
503 out which slabs are relevant to a particular load.
504 Try running: slabinfo -DA
506 config HAVE_DEBUG_KMEMLEAK
509 config DEBUG_KMEMLEAK
510 bool "Kernel memory leak detector"
511 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
513 select STACKTRACE if STACKTRACE_SUPPORT
517 Say Y here if you want to enable the memory leak
518 detector. The memory allocation/freeing is traced in a way
519 similar to the Boehm's conservative garbage collector, the
520 difference being that the orphan objects are not freed but
521 only shown in /sys/kernel/debug/kmemleak. Enabling this
522 feature will introduce an overhead to memory
523 allocations. See Documentation/kmemleak.txt for more
526 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
527 of finding leaks due to the slab objects poisoning.
529 In order to access the kmemleak file, debugfs needs to be
530 mounted (usually at /sys/kernel/debug).
532 config DEBUG_KMEMLEAK_EARLY_LOG_SIZE
533 int "Maximum kmemleak early log entries"
534 depends on DEBUG_KMEMLEAK
538 Kmemleak must track all the memory allocations to avoid
539 reporting false positives. Since memory may be allocated or
540 freed before kmemleak is initialised, an early log buffer is
541 used to store these actions. If kmemleak reports "early log
542 buffer exceeded", please increase this value.
544 config DEBUG_KMEMLEAK_TEST
545 tristate "Simple test for the kernel memory leak detector"
546 depends on DEBUG_KMEMLEAK && m
548 This option enables a module that explicitly leaks memory.
552 config DEBUG_KMEMLEAK_DEFAULT_OFF
553 bool "Default kmemleak to off"
554 depends on DEBUG_KMEMLEAK
556 Say Y here to disable kmemleak by default. It can then be enabled
557 on the command line via kmemleak=on.
559 config DEBUG_STACK_USAGE
560 bool "Stack utilization instrumentation"
561 depends on DEBUG_KERNEL && !IA64
563 Enables the display of the minimum amount of free stack which each
564 task has ever had available in the sysrq-T and sysrq-P debug output.
566 This option will slow down process creation somewhat.
570 depends on DEBUG_KERNEL
572 Enable this to turn on extended checks in the virtual-memory system
573 that may impact performance.
577 config DEBUG_VM_VMACACHE
578 bool "Debug VMA caching"
581 Enable this to turn on VMA caching debug information. Doing so
582 can cause significant overhead, so only enable it in non-production
588 bool "Debug VM red-black trees"
591 Enable VM red-black tree debugging information and extra validations.
595 config DEBUG_VM_PGFLAGS
596 bool "Debug page-flags operations"
599 Enables extra validation on page flags operations.
604 bool "Debug VM translations"
605 depends on DEBUG_KERNEL && X86
607 Enable some costly sanity checks in virtual to page code. This can
608 catch mistakes with virt_to_page() and friends.
612 config DEBUG_NOMMU_REGIONS
613 bool "Debug the global anon/private NOMMU mapping region tree"
614 depends on DEBUG_KERNEL && !MMU
616 This option causes the global tree of anonymous and private mapping
617 regions to be regularly checked for invalid topology.
619 config DEBUG_MEMORY_INIT
620 bool "Debug memory initialisation" if EXPERT
623 Enable this for additional checks during memory initialisation.
624 The sanity checks verify aspects of the VM such as the memory model
625 and other information provided by the architecture. Verbose
626 information will be printed at KERN_DEBUG loglevel depending
627 on the mminit_loglevel= command-line option.
631 config MEMORY_NOTIFIER_ERROR_INJECT
632 tristate "Memory hotplug notifier error injection module"
633 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION
635 This option provides the ability to inject artificial errors to
636 memory hotplug notifier chain callbacks. It is controlled through
637 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
639 If the notifier call chain should be failed with some events
640 notified, write the error code to "actions/<notifier event>/error".
642 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
644 # cd /sys/kernel/debug/notifier-error-inject/memory
645 # echo -12 > actions/MEM_GOING_OFFLINE/error
646 # echo offline > /sys/devices/system/memory/memoryXXX/state
647 bash: echo: write error: Cannot allocate memory
649 To compile this code as a module, choose M here: the module will
650 be called memory-notifier-error-inject.
654 config DEBUG_PER_CPU_MAPS
655 bool "Debug access to per_cpu maps"
656 depends on DEBUG_KERNEL
659 Say Y to verify that the per_cpu map being accessed has
660 been set up. This adds a fair amount of code to kernel memory
661 and decreases performance.
666 bool "Highmem debugging"
667 depends on DEBUG_KERNEL && HIGHMEM
669 This option enables additional error checking for high memory
670 systems. Disable for production systems.
672 config HAVE_DEBUG_STACKOVERFLOW
675 config DEBUG_STACKOVERFLOW
676 bool "Check for stack overflows"
677 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
679 Say Y here if you want to check for overflows of kernel, IRQ
680 and exception stacks (if your architecture uses them). This
681 option will show detailed messages if free stack space drops
682 below a certain limit.
684 These kinds of bugs usually occur when call-chains in the
685 kernel get too deep, especially when interrupts are
688 Use this in cases where you see apparently random memory
689 corruption, especially if it appears in 'struct thread_info'
691 If in doubt, say "N".
693 source "lib/Kconfig.kmemcheck"
695 source "lib/Kconfig.kasan"
697 endmenu # "Memory Debugging"
702 KCOV does not have any arch-specific code, but currently it is enabled
703 only for x86_64. KCOV requires testing on other archs, and most likely
704 disabling of instrumentation for some early boot code.
707 bool "Code coverage for fuzzing"
708 depends on ARCH_HAS_KCOV
711 KCOV exposes kernel code coverage information in a form suitable
712 for coverage-guided fuzzing (randomized testing).
714 If RANDOMIZE_BASE is enabled, PC values will not be stable across
715 different machines and across reboots. If you need stable PC values,
716 disable RANDOMIZE_BASE.
718 For more details, see Documentation/kcov.txt.
721 bool "Debug shared IRQ handlers"
722 depends on DEBUG_KERNEL
724 Enable this to generate a spurious interrupt as soon as a shared
725 interrupt handler is registered, and just before one is deregistered.
726 Drivers ought to be able to handle interrupts coming in at those
727 points; some don't and need to be caught.
729 menu "Debug Lockups and Hangs"
731 config LOCKUP_DETECTOR
732 bool "Detect Hard and Soft Lockups"
733 depends on DEBUG_KERNEL && !S390
735 Say Y here to enable the kernel to act as a watchdog to detect
736 hard and soft lockups.
738 Softlockups are bugs that cause the kernel to loop in kernel
739 mode for more than 20 seconds, without giving other tasks a
740 chance to run. The current stack trace is displayed upon
741 detection and the system will stay locked up.
743 Hardlockups are bugs that cause the CPU to loop in kernel mode
744 for more than 10 seconds, without letting other interrupts have a
745 chance to run. The current stack trace is displayed upon detection
746 and the system will stay locked up.
748 The overhead should be minimal. A periodic hrtimer runs to
749 generate interrupts and kick the watchdog task every 4 seconds.
750 An NMI is generated every 10 seconds or so to check for hardlockups.
752 The frequency of hrtimer and NMI events and the soft and hard lockup
753 thresholds can be controlled through the sysctl watchdog_thresh.
755 config HARDLOCKUP_DETECTOR
757 depends on LOCKUP_DETECTOR && !HAVE_NMI_WATCHDOG
758 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI
760 config BOOTPARAM_HARDLOCKUP_PANIC
761 bool "Panic (Reboot) On Hard Lockups"
762 depends on HARDLOCKUP_DETECTOR
764 Say Y here to enable the kernel to panic on "hard lockups",
765 which are bugs that cause the kernel to loop in kernel
766 mode with interrupts disabled for more than 10 seconds (configurable
767 using the watchdog_thresh sysctl).
771 config BOOTPARAM_HARDLOCKUP_PANIC_VALUE
773 depends on HARDLOCKUP_DETECTOR
775 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC
776 default 1 if BOOTPARAM_HARDLOCKUP_PANIC
778 config BOOTPARAM_SOFTLOCKUP_PANIC
779 bool "Panic (Reboot) On Soft Lockups"
780 depends on LOCKUP_DETECTOR
782 Say Y here to enable the kernel to panic on "soft lockups",
783 which are bugs that cause the kernel to loop in kernel
784 mode for more than 20 seconds (configurable using the watchdog_thresh
785 sysctl), without giving other tasks a chance to run.
787 The panic can be used in combination with panic_timeout,
788 to cause the system to reboot automatically after a
789 lockup has been detected. This feature is useful for
790 high-availability systems that have uptime guarantees and
791 where a lockup must be resolved ASAP.
795 config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
797 depends on LOCKUP_DETECTOR
799 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
800 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
802 config DETECT_HUNG_TASK
803 bool "Detect Hung Tasks"
804 depends on DEBUG_KERNEL
805 default LOCKUP_DETECTOR
807 Say Y here to enable the kernel to detect "hung tasks",
808 which are bugs that cause the task to be stuck in
809 uninterruptible "D" state indefinitiley.
811 When a hung task is detected, the kernel will print the
812 current stack trace (which you should report), but the
813 task will stay in uninterruptible state. If lockdep is
814 enabled then all held locks will also be reported. This
815 feature has negligible overhead.
817 config DEFAULT_HUNG_TASK_TIMEOUT
818 int "Default timeout for hung task detection (in seconds)"
819 depends on DETECT_HUNG_TASK
822 This option controls the default timeout (in seconds) used
823 to determine when a task has become non-responsive and should
826 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
827 sysctl or by writing a value to
828 /proc/sys/kernel/hung_task_timeout_secs.
830 A timeout of 0 disables the check. The default is two minutes.
831 Keeping the default should be fine in most cases.
833 config BOOTPARAM_HUNG_TASK_PANIC
834 bool "Panic (Reboot) On Hung Tasks"
835 depends on DETECT_HUNG_TASK
837 Say Y here to enable the kernel to panic on "hung tasks",
838 which are bugs that cause the kernel to leave a task stuck
839 in uninterruptible "D" state.
841 The panic can be used in combination with panic_timeout,
842 to cause the system to reboot automatically after a
843 hung task has been detected. This feature is useful for
844 high-availability systems that have uptime guarantees and
845 where a hung tasks must be resolved ASAP.
849 config BOOTPARAM_HUNG_TASK_PANIC_VALUE
851 depends on DETECT_HUNG_TASK
853 default 0 if !BOOTPARAM_HUNG_TASK_PANIC
854 default 1 if BOOTPARAM_HUNG_TASK_PANIC
857 bool "Detect Workqueue Stalls"
858 depends on DEBUG_KERNEL
860 Say Y here to enable stall detection on workqueues. If a
861 worker pool doesn't make forward progress on a pending work
862 item for over a given amount of time, 30s by default, a
863 warning message is printed along with dump of workqueue
864 state. This can be configured through kernel parameter
865 "workqueue.watchdog_thresh" and its sysfs counterpart.
867 endmenu # "Debug lockups and hangs"
872 Say Y here to enable the kernel to panic when it oopses. This
873 has the same effect as setting oops=panic on the kernel command
876 This feature is useful to ensure that the kernel does not do
877 anything erroneous after an oops which could result in data
878 corruption or other issues.
882 config PANIC_ON_OOPS_VALUE
885 default 0 if !PANIC_ON_OOPS
886 default 1 if PANIC_ON_OOPS
892 Set the timeout value (in seconds) until a reboot occurs when the
893 the kernel panics. If n = 0, then we wait forever. A timeout
894 value n > 0 will wait n seconds before rebooting, while a timeout
895 value n < 0 will reboot immediately.
898 bool "Collect scheduler debugging info"
899 depends on DEBUG_KERNEL && PROC_FS
902 If you say Y here, the /proc/sched_debug file will be provided
903 that can help debug the scheduler. The runtime overhead of this
911 bool "Collect scheduler statistics"
912 depends on DEBUG_KERNEL && PROC_FS
915 If you say Y here, additional code will be inserted into the
916 scheduler and related routines to collect statistics about
917 scheduler behavior and provide them in /proc/schedstat. These
918 stats may be useful for both tuning and debugging the scheduler
919 If you aren't debugging the scheduler or trying to tune a specific
920 application, you can say N to avoid the very slight overhead
923 config SCHED_STACK_END_CHECK
924 bool "Detect stack corruption on calls to schedule()"
925 depends on DEBUG_KERNEL
928 This option checks for a stack overrun on calls to schedule().
929 If the stack end location is found to be over written always panic as
930 the content of the corrupted region can no longer be trusted.
931 This is to ensure no erroneous behaviour occurs which could result in
932 data corruption or a sporadic crash at a later stage once the region
933 is examined. The runtime overhead introduced is minimal.
935 config DEBUG_TIMEKEEPING
936 bool "Enable extra timekeeping sanity checking"
938 This option will enable additional timekeeping sanity checks
939 which may be helpful when diagnosing issues where timekeeping
940 problems are suspected.
942 This may include checks in the timekeeping hotpaths, so this
943 option may have a (very small) performance impact to some
949 bool "Collect kernel timers statistics"
950 depends on DEBUG_KERNEL && PROC_FS
952 If you say Y here, additional code will be inserted into the
953 timer routines to collect statistics about kernel timers being
954 reprogrammed. The statistics can be read from /proc/timer_stats.
955 The statistics collection is started by writing 1 to /proc/timer_stats,
956 writing 0 stops it. This feature is useful to collect information
957 about timer usage patterns in kernel and userspace. This feature
958 is lightweight if enabled in the kernel config but not activated
959 (it defaults to deactivated on bootup and will only be activated
960 if some application like powertop activates it explicitly).
963 bool "Debug preemptible kernel"
964 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT
967 If you say Y here then the kernel will use a debug variant of the
968 commonly used smp_processor_id() function and will print warnings
969 if kernel code uses it in a preemption-unsafe way. Also, the kernel
970 will detect preemption count underflows.
972 menu "Lock Debugging (spinlocks, mutexes, etc...)"
974 config DEBUG_RT_MUTEXES
975 bool "RT Mutex debugging, deadlock detection"
976 depends on DEBUG_KERNEL && RT_MUTEXES
978 This allows rt mutex semantics violations and rt mutex related
979 deadlocks (lockups) to be detected and reported automatically.
981 config DEBUG_SPINLOCK
982 bool "Spinlock and rw-lock debugging: basic checks"
983 depends on DEBUG_KERNEL
984 select UNINLINE_SPIN_UNLOCK
986 Say Y here and build SMP to catch missing spinlock initialization
987 and certain other kinds of spinlock errors commonly made. This is
988 best used in conjunction with the NMI watchdog so that spinlock
989 deadlocks are also debuggable.
992 bool "Mutex debugging: basic checks"
993 depends on DEBUG_KERNEL
995 This feature allows mutex semantics violations to be detected and
998 config DEBUG_WW_MUTEX_SLOWPATH
999 bool "Wait/wound mutex debugging: Slowpath testing"
1000 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1001 select DEBUG_LOCK_ALLOC
1002 select DEBUG_SPINLOCK
1003 select DEBUG_MUTEXES
1005 This feature enables slowpath testing for w/w mutex users by
1006 injecting additional -EDEADLK wound/backoff cases. Together with
1007 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1008 will test all possible w/w mutex interface abuse with the
1009 exception of simply not acquiring all the required locks.
1010 Note that this feature can introduce significant overhead, so
1011 it really should not be enabled in a production or distro kernel,
1012 even a debug kernel. If you are a driver writer, enable it. If
1013 you are a distro, do not.
1015 config DEBUG_LOCK_ALLOC
1016 bool "Lock debugging: detect incorrect freeing of live locks"
1017 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1018 select DEBUG_SPINLOCK
1019 select DEBUG_MUTEXES
1022 This feature will check whether any held lock (spinlock, rwlock,
1023 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1024 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1025 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1026 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1027 held during task exit.
1029 config PROVE_LOCKING
1030 bool "Lock debugging: prove locking correctness"
1031 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1033 select DEBUG_SPINLOCK
1034 select DEBUG_MUTEXES
1035 select DEBUG_LOCK_ALLOC
1036 select TRACE_IRQFLAGS
1039 This feature enables the kernel to prove that all locking
1040 that occurs in the kernel runtime is mathematically
1041 correct: that under no circumstance could an arbitrary (and
1042 not yet triggered) combination of observed locking
1043 sequences (on an arbitrary number of CPUs, running an
1044 arbitrary number of tasks and interrupt contexts) cause a
1047 In short, this feature enables the kernel to report locking
1048 related deadlocks before they actually occur.
1050 The proof does not depend on how hard and complex a
1051 deadlock scenario would be to trigger: how many
1052 participant CPUs, tasks and irq-contexts would be needed
1053 for it to trigger. The proof also does not depend on
1054 timing: if a race and a resulting deadlock is possible
1055 theoretically (no matter how unlikely the race scenario
1056 is), it will be proven so and will immediately be
1057 reported by the kernel (once the event is observed that
1058 makes the deadlock theoretically possible).
1060 If a deadlock is impossible (i.e. the locking rules, as
1061 observed by the kernel, are mathematically correct), the
1062 kernel reports nothing.
1064 NOTE: this feature can also be enabled for rwlocks, mutexes
1065 and rwsems - in which case all dependencies between these
1066 different locking variants are observed and mapped too, and
1067 the proof of observed correctness is also maintained for an
1068 arbitrary combination of these separate locking variants.
1070 For more details, see Documentation/locking/lockdep-design.txt.
1074 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1076 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE && !ARC && !SCORE
1081 bool "Lock usage statistics"
1082 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1084 select DEBUG_SPINLOCK
1085 select DEBUG_MUTEXES
1086 select DEBUG_LOCK_ALLOC
1089 This feature enables tracking lock contention points
1091 For more details, see Documentation/locking/lockstat.txt
1093 This also enables lock events required by "perf lock",
1095 If you want to use "perf lock", you also need to turn on
1096 CONFIG_EVENT_TRACING.
1098 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1099 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1101 config DEBUG_LOCKDEP
1102 bool "Lock dependency engine debugging"
1103 depends on DEBUG_KERNEL && LOCKDEP
1105 If you say Y here, the lock dependency engine will do
1106 additional runtime checks to debug itself, at the price
1107 of more runtime overhead.
1109 config DEBUG_ATOMIC_SLEEP
1110 bool "Sleep inside atomic section checking"
1111 select PREEMPT_COUNT
1112 depends on DEBUG_KERNEL
1114 If you say Y here, various routines which may sleep will become very
1115 noisy if they are called inside atomic sections: when a spinlock is
1116 held, inside an rcu read side critical section, inside preempt disabled
1117 sections, inside an interrupt, etc...
1119 config DEBUG_LOCKING_API_SELFTESTS
1120 bool "Locking API boot-time self-tests"
1121 depends on DEBUG_KERNEL
1123 Say Y here if you want the kernel to run a short self-test during
1124 bootup. The self-test checks whether common types of locking bugs
1125 are detected by debugging mechanisms or not. (if you disable
1126 lock debugging then those bugs wont be detected of course.)
1127 The following locking APIs are covered: spinlocks, rwlocks,
1130 config LOCK_TORTURE_TEST
1131 tristate "torture tests for locking"
1132 depends on DEBUG_KERNEL
1136 This option provides a kernel module that runs torture tests
1137 on kernel locking primitives. The kernel module may be built
1138 after the fact on the running kernel to be tested, if desired.
1140 Say Y here if you want kernel locking-primitive torture tests
1141 to be built into the kernel.
1142 Say M if you want these torture tests to build as a module.
1143 Say N if you are unsure.
1145 endmenu # lock debugging
1147 config TRACE_IRQFLAGS
1150 Enables hooks to interrupt enabling and disabling for
1151 either tracing or lock debugging.
1154 bool "Stack backtrace support"
1155 depends on STACKTRACE_SUPPORT
1157 This option causes the kernel to create a /proc/pid/stack for
1158 every process, showing its current stack trace.
1159 It is also used by various kernel debugging features that require
1160 stack trace generation.
1162 config DEBUG_KOBJECT
1163 bool "kobject debugging"
1164 depends on DEBUG_KERNEL
1166 If you say Y here, some extra kobject debugging messages will be sent
1169 config DEBUG_KOBJECT_RELEASE
1170 bool "kobject release debugging"
1171 depends on DEBUG_OBJECTS_TIMERS
1173 kobjects are reference counted objects. This means that their
1174 last reference count put is not predictable, and the kobject can
1175 live on past the point at which a driver decides to drop it's
1176 initial reference to the kobject gained on allocation. An
1177 example of this would be a struct device which has just been
1180 However, some buggy drivers assume that after such an operation,
1181 the memory backing the kobject can be immediately freed. This
1182 goes completely against the principles of a refcounted object.
1184 If you say Y here, the kernel will delay the release of kobjects
1185 on the last reference count to improve the visibility of this
1186 kind of kobject release bug.
1188 config HAVE_DEBUG_BUGVERBOSE
1191 config DEBUG_BUGVERBOSE
1192 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT
1193 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE)
1196 Say Y here to make BUG() panics output the file name and line number
1197 of the BUG call as well as the EIP and oops trace. This aids
1198 debugging but costs about 70-100K of memory.
1201 bool "Debug linked list manipulation"
1202 depends on DEBUG_KERNEL
1204 Enable this to turn on extended checks in the linked-list
1209 config DEBUG_PI_LIST
1210 bool "Debug priority linked list manipulation"
1211 depends on DEBUG_KERNEL
1213 Enable this to turn on extended checks in the priority-ordered
1214 linked-list (plist) walking routines. This checks the entire
1215 list multiple times during each manipulation.
1220 bool "Debug SG table operations"
1221 depends on DEBUG_KERNEL
1223 Enable this to turn on checks on scatter-gather tables. This can
1224 help find problems with drivers that do not properly initialize
1229 config DEBUG_NOTIFIERS
1230 bool "Debug notifier call chains"
1231 depends on DEBUG_KERNEL
1233 Enable this to turn on sanity checking for notifier call chains.
1234 This is most useful for kernel developers to make sure that
1235 modules properly unregister themselves from notifier chains.
1236 This is a relatively cheap check but if you care about maximum
1239 config DEBUG_CREDENTIALS
1240 bool "Debug credential management"
1241 depends on DEBUG_KERNEL
1243 Enable this to turn on some debug checking for credential
1244 management. The additional code keeps track of the number of
1245 pointers from task_structs to any given cred struct, and checks to
1246 see that this number never exceeds the usage count of the cred
1249 Furthermore, if SELinux is enabled, this also checks that the
1250 security pointer in the cred struct is never seen to be invalid.
1254 menu "RCU Debugging"
1257 def_bool PROVE_LOCKING
1259 config PROVE_RCU_REPEATEDLY
1260 bool "RCU debugging: don't disable PROVE_RCU on first splat"
1261 depends on PROVE_RCU
1264 By itself, PROVE_RCU will disable checking upon issuing the
1265 first warning (or "splat"). This feature prevents such
1266 disabling, allowing multiple RCU-lockdep warnings to be printed
1269 Say Y to allow multiple RCU-lockdep warnings per boot.
1271 Say N if you are unsure.
1273 config SPARSE_RCU_POINTER
1274 bool "RCU debugging: sparse-based checks for pointer usage"
1277 This feature enables the __rcu sparse annotation for
1278 RCU-protected pointers. This annotation will cause sparse
1279 to flag any non-RCU used of annotated pointers. This can be
1280 helpful when debugging RCU usage. Please note that this feature
1281 is not intended to enforce code cleanliness; it is instead merely
1284 Say Y to make sparse flag questionable use of RCU-protected pointers
1286 Say N if you are unsure.
1292 config RCU_TORTURE_TEST
1293 tristate "torture tests for RCU"
1294 depends on DEBUG_KERNEL
1300 This option provides a kernel module that runs torture tests
1301 on the RCU infrastructure. The kernel module may be built
1302 after the fact on the running kernel to be tested, if desired.
1304 Say Y here if you want RCU torture tests to be built into
1306 Say M if you want the RCU torture tests to build as a module.
1307 Say N if you are unsure.
1309 config RCU_TORTURE_TEST_RUNNABLE
1310 bool "torture tests for RCU runnable by default"
1311 depends on RCU_TORTURE_TEST = y
1314 This option provides a way to build the RCU torture tests
1315 directly into the kernel without them starting up at boot
1316 time. You can use /proc/sys/kernel/rcutorture_runnable
1317 to manually override this setting. This /proc file is
1318 available only when the RCU torture tests have been built
1321 Say Y here if you want the RCU torture tests to start during
1322 boot (you probably don't).
1323 Say N here if you want the RCU torture tests to start only
1324 after being manually enabled via /proc.
1326 config RCU_TORTURE_TEST_SLOW_PREINIT
1327 bool "Slow down RCU grace-period pre-initialization to expose races"
1328 depends on RCU_TORTURE_TEST
1330 This option delays grace-period pre-initialization (the
1331 propagation of CPU-hotplug changes up the rcu_node combining
1332 tree) for a few jiffies between initializing each pair of
1333 consecutive rcu_node structures. This helps to expose races
1334 involving grace-period pre-initialization, in other words, it
1335 makes your kernel less stable. It can also greatly increase
1336 grace-period latency, especially on systems with large numbers
1337 of CPUs. This is useful when torture-testing RCU, but in
1338 almost no other circumstance.
1340 Say Y here if you want your system to crash and hang more often.
1341 Say N if you want a sane system.
1343 config RCU_TORTURE_TEST_SLOW_PREINIT_DELAY
1344 int "How much to slow down RCU grace-period pre-initialization"
1347 depends on RCU_TORTURE_TEST_SLOW_PREINIT
1349 This option specifies the number of jiffies to wait between
1350 each rcu_node structure pre-initialization step.
1352 config RCU_TORTURE_TEST_SLOW_INIT
1353 bool "Slow down RCU grace-period initialization to expose races"
1354 depends on RCU_TORTURE_TEST
1356 This option delays grace-period initialization for a few
1357 jiffies between initializing each pair of consecutive
1358 rcu_node structures. This helps to expose races involving
1359 grace-period initialization, in other words, it makes your
1360 kernel less stable. It can also greatly increase grace-period
1361 latency, especially on systems with large numbers of CPUs.
1362 This is useful when torture-testing RCU, but in almost no
1365 Say Y here if you want your system to crash and hang more often.
1366 Say N if you want a sane system.
1368 config RCU_TORTURE_TEST_SLOW_INIT_DELAY
1369 int "How much to slow down RCU grace-period initialization"
1372 depends on RCU_TORTURE_TEST_SLOW_INIT
1374 This option specifies the number of jiffies to wait between
1375 each rcu_node structure initialization.
1377 config RCU_TORTURE_TEST_SLOW_CLEANUP
1378 bool "Slow down RCU grace-period cleanup to expose races"
1379 depends on RCU_TORTURE_TEST
1381 This option delays grace-period cleanup for a few jiffies
1382 between cleaning up each pair of consecutive rcu_node
1383 structures. This helps to expose races involving grace-period
1384 cleanup, in other words, it makes your kernel less stable.
1385 It can also greatly increase grace-period latency, especially
1386 on systems with large numbers of CPUs. This is useful when
1387 torture-testing RCU, but in almost no other circumstance.
1389 Say Y here if you want your system to crash and hang more often.
1390 Say N if you want a sane system.
1392 config RCU_TORTURE_TEST_SLOW_CLEANUP_DELAY
1393 int "How much to slow down RCU grace-period cleanup"
1396 depends on RCU_TORTURE_TEST_SLOW_CLEANUP
1398 This option specifies the number of jiffies to wait between
1399 each rcu_node structure cleanup operation.
1401 config RCU_CPU_STALL_TIMEOUT
1402 int "RCU CPU stall timeout in seconds"
1403 depends on RCU_STALL_COMMON
1407 If a given RCU grace period extends more than the specified
1408 number of seconds, a CPU stall warning is printed. If the
1409 RCU grace period persists, additional CPU stall warnings are
1410 printed at more widely spaced intervals.
1413 bool "Enable tracing for RCU"
1414 depends on DEBUG_KERNEL
1417 This option provides tracing in RCU which presents stats
1418 in debugfs for debugging RCU implementation.
1420 Say Y here if you want to enable RCU tracing
1421 Say N if you are unsure.
1423 config RCU_EQS_DEBUG
1424 bool "Provide debugging asserts for adding NO_HZ support to an arch"
1425 depends on DEBUG_KERNEL
1427 This option provides consistency checks in RCU's handling of
1428 NO_HZ. These checks have proven quite helpful in detecting
1429 bugs in arch-specific NO_HZ code.
1431 Say N here if you need ultimate kernel/user switch latencies
1432 Say Y if you are unsure
1434 endmenu # "RCU Debugging"
1436 config DEBUG_WQ_FORCE_RR_CPU
1437 bool "Force round-robin CPU selection for unbound work items"
1438 depends on DEBUG_KERNEL
1441 Workqueue used to implicitly guarantee that work items queued
1442 without explicit CPU specified are put on the local CPU. This
1443 guarantee is no longer true and while local CPU is still
1444 preferred work items may be put on foreign CPUs. Kernel
1445 parameter "workqueue.debug_force_rr_cpu" is added to force
1446 round-robin CPU selection to flush out usages which depend on the
1447 now broken guarantee. This config option enables the debug
1448 feature by default. When enabled, memory and cache locality will
1451 config DEBUG_BLOCK_EXT_DEVT
1452 bool "Force extended block device numbers and spread them"
1453 depends on DEBUG_KERNEL
1457 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON
1458 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT
1459 YOU ARE DOING. Distros, please enable this and fix whatever
1462 Conventionally, block device numbers are allocated from
1463 predetermined contiguous area. However, extended block area
1464 may introduce non-contiguous block device numbers. This
1465 option forces most block device numbers to be allocated from
1466 the extended space and spreads them to discover kernel or
1467 userland code paths which assume predetermined contiguous
1468 device number allocation.
1470 Note that turning on this debug option shuffles all the
1471 device numbers for all IDE and SCSI devices including libata
1472 ones, so root partition specified using device number
1473 directly (via rdev or root=MAJ:MIN) won't work anymore.
1474 Textual device names (root=/dev/sdXn) will continue to work.
1476 Say N if you are unsure.
1478 config CPU_HOTPLUG_STATE_CONTROL
1479 bool "Enable CPU hotplug state control"
1480 depends on DEBUG_KERNEL
1481 depends on HOTPLUG_CPU
1484 Allows to write steps between "offline" and "online" to the CPUs
1485 sysfs target file so states can be stepped granular. This is a debug
1486 option for now as the hotplug machinery cannot be stopped and
1487 restarted at arbitrary points yet.
1489 Say N if your are unsure.
1491 config NOTIFIER_ERROR_INJECTION
1492 tristate "Notifier error injection"
1493 depends on DEBUG_KERNEL
1496 This option provides the ability to inject artificial errors to
1497 specified notifier chain callbacks. It is useful to test the error
1498 handling of notifier call chain failures.
1502 config CPU_NOTIFIER_ERROR_INJECT
1503 tristate "CPU notifier error injection module"
1504 depends on HOTPLUG_CPU && NOTIFIER_ERROR_INJECTION
1506 This option provides a kernel module that can be used to test
1507 the error handling of the cpu notifiers by injecting artificial
1508 errors to CPU notifier chain callbacks. It is controlled through
1509 debugfs interface under /sys/kernel/debug/notifier-error-inject/cpu
1511 If the notifier call chain should be failed with some events
1512 notified, write the error code to "actions/<notifier event>/error".
1514 Example: Inject CPU offline error (-1 == -EPERM)
1516 # cd /sys/kernel/debug/notifier-error-inject/cpu
1517 # echo -1 > actions/CPU_DOWN_PREPARE/error
1518 # echo 0 > /sys/devices/system/cpu/cpu1/online
1519 bash: echo: write error: Operation not permitted
1521 To compile this code as a module, choose M here: the module will
1522 be called cpu-notifier-error-inject.
1526 config PM_NOTIFIER_ERROR_INJECT
1527 tristate "PM notifier error injection module"
1528 depends on PM && NOTIFIER_ERROR_INJECTION
1529 default m if PM_DEBUG
1531 This option provides the ability to inject artificial errors to
1532 PM notifier chain callbacks. It is controlled through debugfs
1533 interface /sys/kernel/debug/notifier-error-inject/pm
1535 If the notifier call chain should be failed with some events
1536 notified, write the error code to "actions/<notifier event>/error".
1538 Example: Inject PM suspend error (-12 = -ENOMEM)
1540 # cd /sys/kernel/debug/notifier-error-inject/pm/
1541 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1542 # echo mem > /sys/power/state
1543 bash: echo: write error: Cannot allocate memory
1545 To compile this code as a module, choose M here: the module will
1546 be called pm-notifier-error-inject.
1550 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1551 tristate "OF reconfig notifier error injection module"
1552 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1554 This option provides the ability to inject artificial errors to
1555 OF reconfig notifier chain callbacks. It is controlled
1556 through debugfs interface under
1557 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1559 If the notifier call chain should be failed with some events
1560 notified, write the error code to "actions/<notifier event>/error".
1562 To compile this code as a module, choose M here: the module will
1563 be called of-reconfig-notifier-error-inject.
1567 config NETDEV_NOTIFIER_ERROR_INJECT
1568 tristate "Netdev notifier error injection module"
1569 depends on NET && NOTIFIER_ERROR_INJECTION
1571 This option provides the ability to inject artificial errors to
1572 netdevice notifier chain callbacks. It is controlled through debugfs
1573 interface /sys/kernel/debug/notifier-error-inject/netdev
1575 If the notifier call chain should be failed with some events
1576 notified, write the error code to "actions/<notifier event>/error".
1578 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1580 # cd /sys/kernel/debug/notifier-error-inject/netdev
1581 # echo -22 > actions/NETDEV_CHANGEMTU/error
1582 # ip link set eth0 mtu 1024
1583 RTNETLINK answers: Invalid argument
1585 To compile this code as a module, choose M here: the module will
1586 be called netdev-notifier-error-inject.
1590 config FAULT_INJECTION
1591 bool "Fault-injection framework"
1592 depends on DEBUG_KERNEL
1594 Provide fault-injection framework.
1595 For more details, see Documentation/fault-injection/.
1598 bool "Fault-injection capability for kmalloc"
1599 depends on FAULT_INJECTION
1600 depends on SLAB || SLUB
1602 Provide fault-injection capability for kmalloc.
1604 config FAIL_PAGE_ALLOC
1605 bool "Fault-injection capabilitiy for alloc_pages()"
1606 depends on FAULT_INJECTION
1608 Provide fault-injection capability for alloc_pages().
1610 config FAIL_MAKE_REQUEST
1611 bool "Fault-injection capability for disk IO"
1612 depends on FAULT_INJECTION && BLOCK
1614 Provide fault-injection capability for disk IO.
1616 config FAIL_IO_TIMEOUT
1617 bool "Fault-injection capability for faking disk interrupts"
1618 depends on FAULT_INJECTION && BLOCK
1620 Provide fault-injection capability on end IO handling. This
1621 will make the block layer "forget" an interrupt as configured,
1622 thus exercising the error handling.
1624 Only works with drivers that use the generic timeout handling,
1625 for others it wont do anything.
1627 config FAIL_MMC_REQUEST
1628 bool "Fault-injection capability for MMC IO"
1629 depends on FAULT_INJECTION_DEBUG_FS && MMC
1631 Provide fault-injection capability for MMC IO.
1632 This will make the mmc core return data errors. This is
1633 useful to test the error handling in the mmc block device
1634 and to test how the mmc host driver handles retries from
1638 bool "Fault-injection capability for futexes"
1640 depends on FAULT_INJECTION && FUTEX
1642 Provide fault-injection capability for futexes.
1644 config FAULT_INJECTION_DEBUG_FS
1645 bool "Debugfs entries for fault-injection capabilities"
1646 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1648 Enable configuration of fault-injection capabilities via debugfs.
1650 config FAULT_INJECTION_STACKTRACE_FILTER
1651 bool "stacktrace filter for fault-injection capabilities"
1652 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1655 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !SCORE
1657 Provide stacktrace filter for fault-injection capabilities
1660 bool "Latency measuring infrastructure"
1661 depends on DEBUG_KERNEL
1662 depends on STACKTRACE_SUPPORT
1664 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC
1671 Enable this option if you want to use the LatencyTOP tool
1672 to find out which userspace is blocking on what kernel operations.
1674 config ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS
1677 config DEBUG_STRICT_USER_COPY_CHECKS
1678 bool "Strict user copy size checks"
1679 depends on ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS
1680 depends on DEBUG_KERNEL && !TRACE_BRANCH_PROFILING
1682 Enabling this option turns a certain set of sanity checks for user
1683 copy operations into compile time failures.
1685 The copy_from_user() etc checks are there to help test if there
1686 are sufficient security checks on the length argument of
1687 the copy operation, by having gcc prove that the argument is
1692 source kernel/trace/Kconfig
1694 menu "Runtime Testing"
1697 tristate "Linux Kernel Dump Test Tool Module"
1702 This module enables testing of the different dumping mechanisms by
1703 inducing system failures at predefined crash points.
1704 If you don't need it: say N
1705 Choose M here to compile this code as a module. The module will be
1708 Documentation on how to use the module can be found in
1709 Documentation/fault-injection/provoke-crashes.txt
1711 config TEST_LIST_SORT
1712 bool "Linked list sorting test"
1713 depends on DEBUG_KERNEL
1715 Enable this to turn on 'list_sort()' function test. This test is
1716 executed only once during system boot, so affects only boot time.
1720 config KPROBES_SANITY_TEST
1721 bool "Kprobes sanity tests"
1722 depends on DEBUG_KERNEL
1726 This option provides for testing basic kprobes functionality on
1727 boot. A sample kprobe, jprobe and kretprobe are inserted and
1728 verified for functionality.
1730 Say N if you are unsure.
1732 config BACKTRACE_SELF_TEST
1733 tristate "Self test for the backtrace code"
1734 depends on DEBUG_KERNEL
1737 This option provides a kernel module that can be used to test
1738 the kernel stack backtrace code. This option is not useful
1739 for distributions or general kernels, but only for kernel
1740 developers working on architecture code.
1742 Note that if you want to also test saved backtraces, you will
1743 have to enable STACKTRACE as well.
1745 Say N if you are unsure.
1748 tristate "Red-Black tree test"
1749 depends on DEBUG_KERNEL
1751 A benchmark measuring the performance of the rbtree library.
1752 Also includes rbtree invariant checks.
1754 config INTERVAL_TREE_TEST
1755 tristate "Interval tree test"
1756 depends on m && DEBUG_KERNEL
1757 select INTERVAL_TREE
1759 A benchmark measuring the performance of the interval tree library
1762 tristate "Per cpu operations test"
1763 depends on m && DEBUG_KERNEL
1765 Enable this option to build test module which validates per-cpu
1770 config ATOMIC64_SELFTEST
1771 bool "Perform an atomic64_t self-test at boot"
1773 Enable this option to test the atomic64_t functions at boot.
1777 config ASYNC_RAID6_TEST
1778 tristate "Self test for hardware accelerated raid6 recovery"
1779 depends on ASYNC_RAID6_RECOV
1782 This is a one-shot self test that permutes through the
1783 recovery of all the possible two disk failure scenarios for a
1784 N-disk array. Recovery is performed with the asynchronous
1785 raid6 recovery routines, and will optionally use an offload
1786 engine if one is available.
1791 tristate "Test functions located in the hexdump module at runtime"
1793 config TEST_STRING_HELPERS
1794 tristate "Test functions located in the string_helpers module at runtime"
1797 tristate "Test kstrto*() family of functions at runtime"
1800 tristate "Test printf() family of functions at runtime"
1803 tristate "Test bitmap_*() family of functions at runtime"
1806 Enable this option to test the bitmap functions at boot.
1810 config TEST_RHASHTABLE
1811 tristate "Perform selftest on resizable hash table"
1814 Enable this option to test the rhashtable functions at boot.
1818 endmenu # runtime tests
1820 config PROVIDE_OHCI1394_DMA_INIT
1821 bool "Remote debugging over FireWire early on boot"
1822 depends on PCI && X86
1824 If you want to debug problems which hang or crash the kernel early
1825 on boot and the crashing machine has a FireWire port, you can use
1826 this feature to remotely access the memory of the crashed machine
1827 over FireWire. This employs remote DMA as part of the OHCI1394
1828 specification which is now the standard for FireWire controllers.
1830 With remote DMA, you can monitor the printk buffer remotely using
1831 firescope and access all memory below 4GB using fireproxy from gdb.
1832 Even controlling a kernel debugger is possible using remote DMA.
1836 If ohci1394_dma=early is used as boot parameter, it will initialize
1837 all OHCI1394 controllers which are found in the PCI config space.
1839 As all changes to the FireWire bus such as enabling and disabling
1840 devices cause a bus reset and thereby disable remote DMA for all
1841 devices, be sure to have the cable plugged and FireWire enabled on
1842 the debugging host before booting the debug target for debugging.
1844 This code (~1k) is freed after boot. By then, the firewire stack
1845 in charge of the OHCI-1394 controllers should be used instead.
1847 See Documentation/debugging-via-ohci1394.txt for more information.
1850 bool "Build targets in Documentation/ tree"
1851 depends on HEADERS_CHECK
1853 This option attempts to build objects from the source files in the
1854 kernel Documentation/ tree.
1856 Say N if you are unsure.
1858 config DMA_API_DEBUG
1859 bool "Enable debugging of DMA-API usage"
1860 depends on HAVE_DMA_API_DEBUG
1862 Enable this option to debug the use of the DMA API by device drivers.
1863 With this option you will be able to detect common bugs in device
1864 drivers like double-freeing of DMA mappings or freeing mappings that
1865 were never allocated.
1867 This also attempts to catch cases where a page owned by DMA is
1868 accessed by the cpu in a way that could cause data corruption. For
1869 example, this enables cow_user_page() to check that the source page is
1872 This option causes a performance degradation. Use only if you want to
1873 debug device drivers and dma interactions.
1878 tristate "Test module loading with 'hello world' module"
1882 This builds the "test_module" module that emits "Hello, world"
1883 on printk when loaded. It is designed to be used for basic
1884 evaluation of the module loading subsystem (for example when
1885 validating module verification). It lacks any extra dependencies,
1886 and will not normally be loaded by the system unless explicitly
1891 config TEST_USER_COPY
1892 tristate "Test user/kernel boundary protections"
1896 This builds the "test_user_copy" module that runs sanity checks
1897 on the copy_to/from_user infrastructure, making sure basic
1898 user/kernel boundary testing is working. If it fails to load,
1899 a regression has been detected in the user/kernel memory boundary
1905 tristate "Test BPF filter functionality"
1909 This builds the "test_bpf" module that runs various test vectors
1910 against the BPF interpreter or BPF JIT compiler depending on the
1911 current setting. This is in particular useful for BPF JIT compiler
1912 development, but also to run regression tests against changes in
1913 the interpreter code. It also enables test stubs for eBPF maps and
1914 verifier used by user space verifier testsuite.
1918 config TEST_FIRMWARE
1919 tristate "Test firmware loading via userspace interface"
1921 depends on FW_LOADER
1923 This builds the "test_firmware" module that creates a userspace
1924 interface for testing firmware loading. This can be used to
1925 control the triggering of firmware loading without needing an
1926 actual firmware-using device. The contents can be rechecked by
1932 tristate "udelay test driver"
1935 This builds the "udelay_test" module that helps to make sure
1936 that udelay() is working properly.
1942 depends on HAVE_MEMBLOCK
1944 This option adds a kernel parameter 'memtest', which allows memtest
1946 memtest=0, mean disabled; -- default
1947 memtest=1, mean do 1 test pattern;
1949 memtest=17, mean do 17 test patterns.
1950 If you are unsure how to answer this question, answer N.
1952 config TEST_STATIC_KEYS
1953 tristate "Test static keys"
1957 Test the static key interfaces.
1961 source "samples/Kconfig"
1963 source "lib/Kconfig.kgdb"
1965 source "lib/Kconfig.ubsan"
1967 config ARCH_HAS_DEVMEM_IS_ALLOWED
1970 config STRICT_DEVMEM
1971 bool "Filter access to /dev/mem"
1973 depends on ARCH_HAS_DEVMEM_IS_ALLOWED
1974 default y if TILE || PPC
1976 If this option is disabled, you allow userspace (root) access to all
1977 of memory, including kernel and userspace memory. Accidental
1978 access to this is obviously disastrous, but specific access can
1979 be used by people debugging the kernel. Note that with PAT support
1980 enabled, even in this case there are restrictions on /dev/mem
1981 use due to the cache aliasing requirements.
1983 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
1984 file only allows userspace access to PCI space and the BIOS code and
1985 data regions. This is sufficient for dosemu and X and all common
1990 config IO_STRICT_DEVMEM
1991 bool "Filter I/O access to /dev/mem"
1992 depends on STRICT_DEVMEM
1994 If this option is disabled, you allow userspace (root) access to all
1995 io-memory regardless of whether a driver is actively using that
1996 range. Accidental access to this is obviously disastrous, but
1997 specific access can be used by people debugging kernel drivers.
1999 If this option is switched on, the /dev/mem file only allows
2000 userspace access to *idle* io-memory ranges (see /proc/iomem) This
2001 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
2002 if the driver using a given range cannot be disabled.