3 bool "Show timing information on printks"
6 Selecting this option causes time stamps of the printk()
7 messages to be added to the output of the syslog() system
8 call and at the console.
10 The timestamp is always recorded internally, and exported
11 to /dev/kmsg. This flag just specifies if the timestamp should
12 be included, not that the timestamp is recorded.
14 The behavior is also controlled by the kernel command line
15 parameter printk.time=1. See Documentation/kernel-parameters.txt
18 bool "Show process information on printks"
21 Selecting this option causes process to be
22 included in printk output. Or add printk.process=1 at boot-time.
24 config DEFAULT_MESSAGE_LOGLEVEL
25 int "Default message log level (1-7)"
29 Default log level for printk statements with no specified priority.
31 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks
32 that are auditing their logs closely may want to set it to a lower
35 config ENABLE_WARN_DEPRECATED
36 bool "Enable __deprecated logic"
39 Enable the __deprecated logic in the kernel build.
40 Disable this to suppress the "warning: 'foo' is deprecated
41 (declared at kernel/power/somefile.c:1234)" messages.
43 config ENABLE_MUST_CHECK
44 bool "Enable __must_check logic"
47 Enable the __must_check logic in the kernel build. Disable this to
48 suppress the "warning: ignoring return value of 'foo', declared with
49 attribute warn_unused_result" messages.
52 int "Warn for stack frames larger than (needs gcc 4.4)"
54 default 1024 if !64BIT
57 Tell gcc to warn at build time for stack frames larger than this.
58 Setting this too low will cause a lot of warnings.
59 Setting it to 0 disables the warning.
63 bool "Magic SysRq key"
66 If you say Y here, you will have some control over the system even
67 if the system crashes for example during kernel debugging (e.g., you
68 will be able to flush the buffer cache to disk, reboot the system
69 immediately or dump some status information). This is accomplished
70 by pressing various keys while holding SysRq (Alt+PrintScreen). It
71 also works on a serial console (on PC hardware at least), if you
72 send a BREAK and then within 5 seconds a command keypress. The
73 keys are documented in <file:Documentation/sysrq.txt>. Don't say Y
74 unless you really know what this hack does.
77 bool "Strip assembler-generated symbols during link"
80 Strip internal assembler-generated symbols during a link (symbols
81 that look like '.Lxxx') so they don't pollute the output of
82 get_wchan() and suchlike.
85 bool "Generate readable assembler code"
86 depends on DEBUG_KERNEL
88 Disable some compiler optimizations that tend to generate human unreadable
89 assembler output. This may make the kernel slightly slower, but it helps
90 to keep kernel developers who have to stare a lot at assembler listings
94 bool "Enable unused/obsolete exported symbols"
97 Unused but exported symbols make the kernel needlessly bigger. For
98 that reason most of these unused exports will soon be removed. This
99 option is provided temporarily to provide a transition period in case
100 some external kernel module needs one of these symbols anyway. If you
101 encounter such a case in your module, consider if you are actually
102 using the right API. (rationale: since nobody in the kernel is using
103 this in a module, there is a pretty good chance it's actually the
104 wrong interface to use). If you really need the symbol, please send a
105 mail to the linux kernel mailing list mentioning the symbol and why
106 you really need it, and what the merge plan to the mainline kernel for
110 bool "Debug Filesystem"
112 debugfs is a virtual file system that kernel developers use to put
113 debugging files into. Enable this option to be able to read and
114 write to these files.
116 For detailed documentation on the debugfs API, see
117 Documentation/DocBook/filesystems.
122 bool "Run 'make headers_check' when building vmlinux"
125 This option will extract the user-visible kernel headers whenever
126 building the kernel, and will run basic sanity checks on them to
127 ensure that exported files do not attempt to include files which
128 were not exported, etc.
130 If you're making modifications to header files which are
131 relevant for userspace, say 'Y', and check the headers
132 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in
133 your build tree), to make sure they're suitable.
135 config DEBUG_SECTION_MISMATCH
136 bool "Enable full Section mismatch analysis"
138 The section mismatch analysis checks if there are illegal
139 references from one section to another section.
140 During linktime or runtime, some sections are dropped;
141 any use of code/data previously in these sections would
142 most likely result in an oops.
143 In the code, functions and variables are annotated with
144 __init, __cpuinit, etc. (see the full list in include/linux/init.h),
145 which results in the code/data being placed in specific sections.
146 The section mismatch analysis is always performed after a full
147 kernel build, and enabling this option causes the following
148 additional steps to occur:
149 - Add the option -fno-inline-functions-called-once to gcc commands.
150 When inlining a function annotated with __init in a non-init
151 function, we would lose the section information and thus
152 the analysis would not catch the illegal reference.
153 This option tells gcc to inline less (but it does result in
155 - Run the section mismatch analysis for each module/built-in.o file.
156 When we run the section mismatch analysis on vmlinux.o, we
157 lose valueble information about where the mismatch was
159 Running the analysis for each module/built-in.o file
160 tells where the mismatch happens much closer to the
161 source. The drawback is that the same mismatch is
162 reported at least twice.
163 - Enable verbose reporting from modpost in order to help resolve
164 the section mismatches that are reported.
167 bool "Kernel debugging"
169 Say Y here if you are developing drivers or trying to debug and
170 identify kernel problems.
173 bool "Debug shared IRQ handlers"
174 depends on DEBUG_KERNEL && GENERIC_HARDIRQS
176 Enable this to generate a spurious interrupt as soon as a shared
177 interrupt handler is registered, and just before one is deregistered.
178 Drivers ought to be able to handle interrupts coming in at those
179 points; some don't and need to be caught.
181 config LOCKUP_DETECTOR
182 bool "Detect Hard and Soft Lockups"
183 depends on DEBUG_KERNEL && !S390
185 Say Y here to enable the kernel to act as a watchdog to detect
186 hard and soft lockups.
188 Softlockups are bugs that cause the kernel to loop in kernel
189 mode for more than 20 seconds, without giving other tasks a
190 chance to run. The current stack trace is displayed upon
191 detection and the system will stay locked up.
193 Hardlockups are bugs that cause the CPU to loop in kernel mode
194 for more than 10 seconds, without letting other interrupts have a
195 chance to run. The current stack trace is displayed upon detection
196 and the system will stay locked up.
198 The overhead should be minimal. A periodic hrtimer runs to
199 generate interrupts and kick the watchdog task every 4 seconds.
200 An NMI is generated every 10 seconds or so to check for hardlockups.
201 If NMIs are not available on the platform, every 12 seconds the
202 hrtimer interrupt on one cpu will be used to check for hardlockups
205 The frequency of hrtimer and NMI events and the soft and hard lockup
206 thresholds can be controlled through the sysctl watchdog_thresh.
208 config HARDLOCKUP_DETECTOR_NMI
210 depends on LOCKUP_DETECTOR && !HAVE_NMI_WATCHDOG
211 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI
213 config HARDLOCKUP_DETECTOR_OTHER_CPU
215 depends on LOCKUP_DETECTOR && SMP
216 depends on !HARDLOCKUP_DETECTOR_NMI && !HAVE_NMI_WATCHDOG
218 config HARDLOCKUP_DETECTOR
220 depends on HARDLOCKUP_DETECTOR_NMI || HARDLOCKUP_DETECTOR_OTHER_CPU
222 config BOOTPARAM_HARDLOCKUP_PANIC
223 bool "Panic (Reboot) On Hard Lockups"
224 depends on HARDLOCKUP_DETECTOR
226 Say Y here to enable the kernel to panic on "hard lockups",
227 which are bugs that cause the kernel to loop in kernel
228 mode with interrupts disabled for more than 10 seconds (configurable
229 using the watchdog_thresh sysctl).
233 config BOOTPARAM_HARDLOCKUP_PANIC_VALUE
235 depends on HARDLOCKUP_DETECTOR
237 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC
238 default 1 if BOOTPARAM_HARDLOCKUP_PANIC
240 config BOOTPARAM_SOFTLOCKUP_PANIC
241 bool "Panic (Reboot) On Soft Lockups"
242 depends on LOCKUP_DETECTOR
244 Say Y here to enable the kernel to panic on "soft lockups",
245 which are bugs that cause the kernel to loop in kernel
246 mode for more than 20 seconds (configurable using the watchdog_thresh
247 sysctl), without giving other tasks a chance to run.
249 The panic can be used in combination with panic_timeout,
250 to cause the system to reboot automatically after a
251 lockup has been detected. This feature is useful for
252 high-availability systems that have uptime guarantees and
253 where a lockup must be resolved ASAP.
257 config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
259 depends on LOCKUP_DETECTOR
261 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
262 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
267 Say Y here to enable the kernel to panic when it oopses. This
268 has the same effect as setting oops=panic on the kernel command
271 This feature is useful to ensure that the kernel does not do
272 anything erroneous after an oops which could result in data
273 corruption or other issues.
277 config PANIC_ON_OOPS_VALUE
280 default 0 if !PANIC_ON_OOPS
281 default 1 if PANIC_ON_OOPS
283 config DETECT_HUNG_TASK
284 bool "Detect Hung Tasks"
285 depends on DEBUG_KERNEL
286 default LOCKUP_DETECTOR
288 Say Y here to enable the kernel to detect "hung tasks",
289 which are bugs that cause the task to be stuck in
290 uninterruptible "D" state indefinitiley.
292 When a hung task is detected, the kernel will print the
293 current stack trace (which you should report), but the
294 task will stay in uninterruptible state. If lockdep is
295 enabled then all held locks will also be reported. This
296 feature has negligible overhead.
298 config DEFAULT_HUNG_TASK_TIMEOUT
299 int "Default timeout for hung task detection (in seconds)"
300 depends on DETECT_HUNG_TASK
303 This option controls the default timeout (in seconds) used
304 to determine when a task has become non-responsive and should
307 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
308 sysctl or by writing a value to
309 /proc/sys/kernel/hung_task_timeout_secs.
311 A timeout of 0 disables the check. The default is two minutes.
312 Keeping the default should be fine in most cases.
314 config BOOTPARAM_HUNG_TASK_PANIC
315 bool "Panic (Reboot) On Hung Tasks"
316 depends on DETECT_HUNG_TASK
318 Say Y here to enable the kernel to panic on "hung tasks",
319 which are bugs that cause the kernel to leave a task stuck
320 in uninterruptible "D" state.
322 The panic can be used in combination with panic_timeout,
323 to cause the system to reboot automatically after a
324 hung task has been detected. This feature is useful for
325 high-availability systems that have uptime guarantees and
326 where a hung tasks must be resolved ASAP.
330 config BOOTPARAM_HUNG_TASK_PANIC_VALUE
332 depends on DETECT_HUNG_TASK
334 default 0 if !BOOTPARAM_HUNG_TASK_PANIC
335 default 1 if BOOTPARAM_HUNG_TASK_PANIC
338 bool "Collect scheduler debugging info"
339 depends on DEBUG_KERNEL && PROC_FS
342 If you say Y here, the /proc/sched_debug file will be provided
343 that can help debug the scheduler. The runtime overhead of this
347 bool "Collect scheduler statistics"
348 depends on DEBUG_KERNEL && PROC_FS
350 If you say Y here, additional code will be inserted into the
351 scheduler and related routines to collect statistics about
352 scheduler behavior and provide them in /proc/schedstat. These
353 stats may be useful for both tuning and debugging the scheduler
354 If you aren't debugging the scheduler or trying to tune a specific
355 application, you can say N to avoid the very slight overhead
359 bool "Collect kernel timers statistics"
360 depends on DEBUG_KERNEL && PROC_FS
362 If you say Y here, additional code will be inserted into the
363 timer routines to collect statistics about kernel timers being
364 reprogrammed. The statistics can be read from /proc/timer_stats.
365 The statistics collection is started by writing 1 to /proc/timer_stats,
366 writing 0 stops it. This feature is useful to collect information
367 about timer usage patterns in kernel and userspace. This feature
368 is lightweight if enabled in the kernel config but not activated
369 (it defaults to deactivated on bootup and will only be activated
370 if some application like powertop activates it explicitly).
373 bool "Debug object operations"
374 depends on DEBUG_KERNEL
376 If you say Y here, additional code will be inserted into the
377 kernel to track the life time of various objects and validate
378 the operations on those objects.
380 config DEBUG_OBJECTS_SELFTEST
381 bool "Debug objects selftest"
382 depends on DEBUG_OBJECTS
384 This enables the selftest of the object debug code.
386 config DEBUG_OBJECTS_FREE
387 bool "Debug objects in freed memory"
388 depends on DEBUG_OBJECTS
390 This enables checks whether a k/v free operation frees an area
391 which contains an object which has not been deactivated
392 properly. This can make kmalloc/kfree-intensive workloads
395 config DEBUG_OBJECTS_TIMERS
396 bool "Debug timer objects"
397 depends on DEBUG_OBJECTS
399 If you say Y here, additional code will be inserted into the
400 timer routines to track the life time of timer objects and
401 validate the timer operations.
403 config DEBUG_OBJECTS_WORK
404 bool "Debug work objects"
405 depends on DEBUG_OBJECTS
407 If you say Y here, additional code will be inserted into the
408 work queue routines to track the life time of work objects and
409 validate the work operations.
411 config DEBUG_OBJECTS_RCU_HEAD
412 bool "Debug RCU callbacks objects"
413 depends on DEBUG_OBJECTS
415 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
417 config DEBUG_OBJECTS_PERCPU_COUNTER
418 bool "Debug percpu counter objects"
419 depends on DEBUG_OBJECTS
421 If you say Y here, additional code will be inserted into the
422 percpu counter routines to track the life time of percpu counter
423 objects and validate the percpu counter operations.
425 config DEBUG_OBJECTS_ENABLE_DEFAULT
426 int "debug_objects bootup default value (0-1)"
429 depends on DEBUG_OBJECTS
431 Debug objects boot parameter default value
434 bool "Debug slab memory allocations"
435 depends on DEBUG_KERNEL && SLAB && !KMEMCHECK
437 Say Y here to have the kernel do limited verification on memory
438 allocation as well as poisoning memory on free to catch use of freed
439 memory. This can make kmalloc/kfree-intensive workloads much slower.
441 config DEBUG_SLAB_LEAK
442 bool "Memory leak debugging"
443 depends on DEBUG_SLAB
446 bool "SLUB debugging on by default"
447 depends on SLUB && SLUB_DEBUG && !KMEMCHECK
450 Boot with debugging on by default. SLUB boots by default with
451 the runtime debug capabilities switched off. Enabling this is
452 equivalent to specifying the "slub_debug" parameter on boot.
453 There is no support for more fine grained debug control like
454 possible with slub_debug=xxx. SLUB debugging may be switched
455 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
460 bool "Enable SLUB performance statistics"
461 depends on SLUB && SYSFS
463 SLUB statistics are useful to debug SLUBs allocation behavior in
464 order find ways to optimize the allocator. This should never be
465 enabled for production use since keeping statistics slows down
466 the allocator by a few percentage points. The slabinfo command
467 supports the determination of the most active slabs to figure
468 out which slabs are relevant to a particular load.
469 Try running: slabinfo -DA
471 config HAVE_DEBUG_KMEMLEAK
474 config DEBUG_KMEMLEAK
475 bool "Kernel memory leak detector"
476 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
478 select STACKTRACE if STACKTRACE_SUPPORT
482 Say Y here if you want to enable the memory leak
483 detector. The memory allocation/freeing is traced in a way
484 similar to the Boehm's conservative garbage collector, the
485 difference being that the orphan objects are not freed but
486 only shown in /sys/kernel/debug/kmemleak. Enabling this
487 feature will introduce an overhead to memory
488 allocations. See Documentation/kmemleak.txt for more
491 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
492 of finding leaks due to the slab objects poisoning.
494 In order to access the kmemleak file, debugfs needs to be
495 mounted (usually at /sys/kernel/debug).
497 config DEBUG_KMEMLEAK_EARLY_LOG_SIZE
498 int "Maximum kmemleak early log entries"
499 depends on DEBUG_KMEMLEAK
503 Kmemleak must track all the memory allocations to avoid
504 reporting false positives. Since memory may be allocated or
505 freed before kmemleak is initialised, an early log buffer is
506 used to store these actions. If kmemleak reports "early log
507 buffer exceeded", please increase this value.
509 config DEBUG_KMEMLEAK_TEST
510 tristate "Simple test for the kernel memory leak detector"
511 depends on DEBUG_KMEMLEAK && m
513 This option enables a module that explicitly leaks memory.
517 config DEBUG_KMEMLEAK_DEFAULT_OFF
518 bool "Default kmemleak to off"
519 depends on DEBUG_KMEMLEAK
521 Say Y here to disable kmemleak by default. It can then be enabled
522 on the command line via kmemleak=on.
525 bool "Debug preemptible kernel"
526 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT
529 If you say Y here then the kernel will use a debug variant of the
530 commonly used smp_processor_id() function and will print warnings
531 if kernel code uses it in a preemption-unsafe way. Also, the kernel
532 will detect preemption count underflows.
534 config DEBUG_RT_MUTEXES
535 bool "RT Mutex debugging, deadlock detection"
536 depends on DEBUG_KERNEL && RT_MUTEXES
538 This allows rt mutex semantics violations and rt mutex related
539 deadlocks (lockups) to be detected and reported automatically.
544 depends on DEBUG_RT_MUTEXES
546 config RT_MUTEX_TESTER
547 bool "Built-in scriptable tester for rt-mutexes"
548 depends on DEBUG_KERNEL && RT_MUTEXES
550 This option enables a rt-mutex tester.
552 config DEBUG_SPINLOCK
553 bool "Spinlock and rw-lock debugging: basic checks"
554 depends on DEBUG_KERNEL
555 select UNINLINE_SPIN_UNLOCK
557 Say Y here and build SMP to catch missing spinlock initialization
558 and certain other kinds of spinlock errors commonly made. This is
559 best used in conjunction with the NMI watchdog so that spinlock
560 deadlocks are also debuggable.
563 bool "Mutex debugging: basic checks"
564 depends on DEBUG_KERNEL
566 This feature allows mutex semantics violations to be detected and
569 config DEBUG_LOCK_ALLOC
570 bool "Lock debugging: detect incorrect freeing of live locks"
571 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
572 select DEBUG_SPINLOCK
576 This feature will check whether any held lock (spinlock, rwlock,
577 mutex or rwsem) is incorrectly freed by the kernel, via any of the
578 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
579 vfree(), etc.), whether a live lock is incorrectly reinitialized via
580 spin_lock_init()/mutex_init()/etc., or whether there is any lock
581 held during task exit.
584 bool "Lock debugging: prove locking correctness"
585 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
587 select DEBUG_SPINLOCK
589 select DEBUG_LOCK_ALLOC
590 select TRACE_IRQFLAGS
593 This feature enables the kernel to prove that all locking
594 that occurs in the kernel runtime is mathematically
595 correct: that under no circumstance could an arbitrary (and
596 not yet triggered) combination of observed locking
597 sequences (on an arbitrary number of CPUs, running an
598 arbitrary number of tasks and interrupt contexts) cause a
601 In short, this feature enables the kernel to report locking
602 related deadlocks before they actually occur.
604 The proof does not depend on how hard and complex a
605 deadlock scenario would be to trigger: how many
606 participant CPUs, tasks and irq-contexts would be needed
607 for it to trigger. The proof also does not depend on
608 timing: if a race and a resulting deadlock is possible
609 theoretically (no matter how unlikely the race scenario
610 is), it will be proven so and will immediately be
611 reported by the kernel (once the event is observed that
612 makes the deadlock theoretically possible).
614 If a deadlock is impossible (i.e. the locking rules, as
615 observed by the kernel, are mathematically correct), the
616 kernel reports nothing.
618 NOTE: this feature can also be enabled for rwlocks, mutexes
619 and rwsems - in which case all dependencies between these
620 different locking variants are observed and mapped too, and
621 the proof of observed correctness is also maintained for an
622 arbitrary combination of these separate locking variants.
624 For more details, see Documentation/lockdep-design.txt.
628 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
630 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE
635 bool "Lock usage statistics"
636 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
638 select DEBUG_SPINLOCK
640 select DEBUG_LOCK_ALLOC
643 This feature enables tracking lock contention points
645 For more details, see Documentation/lockstat.txt
647 This also enables lock events required by "perf lock",
649 If you want to use "perf lock", you also need to turn on
650 CONFIG_EVENT_TRACING.
652 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
653 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
656 bool "Lock dependency engine debugging"
657 depends on DEBUG_KERNEL && LOCKDEP
659 If you say Y here, the lock dependency engine will do
660 additional runtime checks to debug itself, at the price
661 of more runtime overhead.
663 config TRACE_IRQFLAGS
666 Enables hooks to interrupt enabling and disabling for
667 either tracing or lock debugging.
669 config DEBUG_ATOMIC_SLEEP
670 bool "Sleep inside atomic section checking"
672 depends on DEBUG_KERNEL
674 If you say Y here, various routines which may sleep will become very
675 noisy if they are called inside atomic sections: when a spinlock is
676 held, inside an rcu read side critical section, inside preempt disabled
677 sections, inside an interrupt, etc...
679 config DEBUG_LOCKING_API_SELFTESTS
680 bool "Locking API boot-time self-tests"
681 depends on DEBUG_KERNEL
683 Say Y here if you want the kernel to run a short self-test during
684 bootup. The self-test checks whether common types of locking bugs
685 are detected by debugging mechanisms or not. (if you disable
686 lock debugging then those bugs wont be detected of course.)
687 The following locking APIs are covered: spinlocks, rwlocks,
692 depends on STACKTRACE_SUPPORT
695 config DEBUG_STACK_USAGE
696 bool "Stack utilization instrumentation"
697 depends on DEBUG_KERNEL && !IA64 && !PARISC && !METAG
699 Enables the display of the minimum amount of free stack which each
700 task has ever had available in the sysrq-T and sysrq-P debug output.
702 This option will slow down process creation somewhat.
705 bool "kobject debugging"
706 depends on DEBUG_KERNEL
708 If you say Y here, some extra kobject debugging messages will be sent
712 bool "Highmem debugging"
713 depends on DEBUG_KERNEL && HIGHMEM
715 This options enables addition error checking for high memory systems.
716 Disable for production systems.
718 config HAVE_DEBUG_BUGVERBOSE
721 config DEBUG_BUGVERBOSE
722 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT
723 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE)
726 Say Y here to make BUG() panics output the file name and line number
727 of the BUG call as well as the EIP and oops trace. This aids
728 debugging but costs about 70-100K of memory.
731 bool "Compile the kernel with debug info"
732 depends on DEBUG_KERNEL
734 If you say Y here the resulting kernel image will include
735 debugging info resulting in a larger kernel image.
736 This adds debug symbols to the kernel and modules (gcc -g), and
737 is needed if you intend to use kernel crashdump or binary object
738 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
739 Say Y here only if you plan to debug the kernel.
743 config DEBUG_INFO_REDUCED
744 bool "Reduce debugging information"
745 depends on DEBUG_INFO
747 If you say Y here gcc is instructed to generate less debugging
748 information for structure types. This means that tools that
749 need full debugging information (like kgdb or systemtap) won't
750 be happy. But if you merely need debugging information to
751 resolve line numbers there is no loss. Advantage is that
752 build directory object sizes shrink dramatically over a full
753 DEBUG_INFO build and compile times are reduced too.
754 Only works with newer gcc versions.
758 depends on DEBUG_KERNEL
760 Enable this to turn on extended checks in the virtual-memory system
761 that may impact performance.
766 bool "Debug VM red-black trees"
769 Enable this to turn on more extended checks in the virtual-memory
770 system that may impact performance.
775 bool "Debug VM translations"
776 depends on DEBUG_KERNEL && X86
778 Enable some costly sanity checks in virtual to page code. This can
779 catch mistakes with virt_to_page() and friends.
783 config DEBUG_NOMMU_REGIONS
784 bool "Debug the global anon/private NOMMU mapping region tree"
785 depends on DEBUG_KERNEL && !MMU
787 This option causes the global tree of anonymous and private mapping
788 regions to be regularly checked for invalid topology.
790 config DEBUG_WRITECOUNT
791 bool "Debug filesystem writers count"
792 depends on DEBUG_KERNEL
794 Enable this to catch wrong use of the writers count in struct
795 vfsmount. This will increase the size of each file struct by
800 config DEBUG_MEMORY_INIT
801 bool "Debug memory initialisation" if EXPERT
804 Enable this for additional checks during memory initialisation.
805 The sanity checks verify aspects of the VM such as the memory model
806 and other information provided by the architecture. Verbose
807 information will be printed at KERN_DEBUG loglevel depending
808 on the mminit_loglevel= command-line option.
813 bool "Debug linked list manipulation"
814 depends on DEBUG_KERNEL
816 Enable this to turn on extended checks in the linked-list
821 config TEST_LIST_SORT
822 bool "Linked list sorting test"
823 depends on DEBUG_KERNEL
825 Enable this to turn on 'list_sort()' function test. This test is
826 executed only once during system boot, so affects only boot time.
831 bool "Debug SG table operations"
832 depends on DEBUG_KERNEL
834 Enable this to turn on checks on scatter-gather tables. This can
835 help find problems with drivers that do not properly initialize
840 config DEBUG_NOTIFIERS
841 bool "Debug notifier call chains"
842 depends on DEBUG_KERNEL
844 Enable this to turn on sanity checking for notifier call chains.
845 This is most useful for kernel developers to make sure that
846 modules properly unregister themselves from notifier chains.
847 This is a relatively cheap check but if you care about maximum
850 config DEBUG_CREDENTIALS
851 bool "Debug credential management"
852 depends on DEBUG_KERNEL
854 Enable this to turn on some debug checking for credential
855 management. The additional code keeps track of the number of
856 pointers from task_structs to any given cred struct, and checks to
857 see that this number never exceeds the usage count of the cred
860 Furthermore, if SELinux is enabled, this also checks that the
861 security pointer in the cred struct is never seen to be invalid.
866 # Select this config option from the architecture Kconfig, if it
867 # is preferred to always offer frame pointers as a config
868 # option on the architecture (regardless of KERNEL_DEBUG):
870 config ARCH_WANT_FRAME_POINTERS
875 bool "Compile the kernel with frame pointers"
876 depends on DEBUG_KERNEL && \
877 (CRIS || M68K || FRV || UML || \
878 AVR32 || SUPERH || BLACKFIN || MN10300 || METAG) || \
879 ARCH_WANT_FRAME_POINTERS
880 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
882 If you say Y here the resulting kernel image will be slightly
883 larger and slower, but it gives very useful debugging information
884 in case of kernel bugs. (precise oopses/stacktraces/warnings)
886 config BOOT_PRINTK_DELAY
887 bool "Delay each boot printk message by N milliseconds"
888 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY
890 This build option allows you to read kernel boot messages
891 by inserting a short delay after each one. The delay is
892 specified in milliseconds on the kernel command line,
893 using "boot_delay=N".
895 It is likely that you would also need to use "lpj=M" to preset
896 the "loops per jiffie" value.
897 See a previous boot log for the "lpj" value to use for your
898 system, and then set "lpj=M" before setting "boot_delay=N".
899 NOTE: Using this option may adversely affect SMP systems.
900 I.e., processors other than the first one may not boot up.
901 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect
902 what it believes to be lockup conditions.
907 bool "RCU debugging: prove RCU correctness"
908 depends on PROVE_LOCKING
911 This feature enables lockdep extensions that check for correct
912 use of RCU APIs. This is currently under development. Say Y
913 if you want to debug RCU usage or help work on the PROVE_RCU
916 Say N if you are unsure.
918 config PROVE_RCU_REPEATEDLY
919 bool "RCU debugging: don't disable PROVE_RCU on first splat"
923 By itself, PROVE_RCU will disable checking upon issuing the
924 first warning (or "splat"). This feature prevents such
925 disabling, allowing multiple RCU-lockdep warnings to be printed
928 Say Y to allow multiple RCU-lockdep warnings per boot.
930 Say N if you are unsure.
932 config PROVE_RCU_DELAY
933 bool "RCU debugging: preemptible RCU race provocation"
934 depends on DEBUG_KERNEL && PREEMPT_RCU
937 There is a class of races that involve an unlikely preemption
938 of __rcu_read_unlock() just after ->rcu_read_lock_nesting has
939 been set to INT_MIN. This feature inserts a delay at that
940 point to increase the probability of these races.
942 Say Y to increase probability of preemption of __rcu_read_unlock().
944 Say N if you are unsure.
946 config SPARSE_RCU_POINTER
947 bool "RCU debugging: sparse-based checks for pointer usage"
950 This feature enables the __rcu sparse annotation for
951 RCU-protected pointers. This annotation will cause sparse
952 to flag any non-RCU used of annotated pointers. This can be
953 helpful when debugging RCU usage. Please note that this feature
954 is not intended to enforce code cleanliness; it is instead merely
957 Say Y to make sparse flag questionable use of RCU-protected pointers
959 Say N if you are unsure.
961 config RCU_TORTURE_TEST
962 tristate "torture tests for RCU"
963 depends on DEBUG_KERNEL
966 This option provides a kernel module that runs torture tests
967 on the RCU infrastructure. The kernel module may be built
968 after the fact on the running kernel to be tested, if desired.
970 Say Y here if you want RCU torture tests to be built into
972 Say M if you want the RCU torture tests to build as a module.
973 Say N if you are unsure.
975 config RCU_TORTURE_TEST_RUNNABLE
976 bool "torture tests for RCU runnable by default"
977 depends on RCU_TORTURE_TEST = y
980 This option provides a way to build the RCU torture tests
981 directly into the kernel without them starting up at boot
982 time. You can use /proc/sys/kernel/rcutorture_runnable
983 to manually override this setting. This /proc file is
984 available only when the RCU torture tests have been built
987 Say Y here if you want the RCU torture tests to start during
988 boot (you probably don't).
989 Say N here if you want the RCU torture tests to start only
990 after being manually enabled via /proc.
992 config RCU_CPU_STALL_TIMEOUT
993 int "RCU CPU stall timeout in seconds"
994 depends on RCU_STALL_COMMON
998 If a given RCU grace period extends more than the specified
999 number of seconds, a CPU stall warning is printed. If the
1000 RCU grace period persists, additional CPU stall warnings are
1001 printed at more widely spaced intervals.
1003 config RCU_CPU_STALL_VERBOSE
1004 bool "Print additional per-task information for RCU_CPU_STALL_DETECTOR"
1005 depends on TREE_PREEMPT_RCU
1008 This option causes RCU to printk detailed per-task information
1009 for any tasks that are stalling the current RCU grace period.
1011 Say N if you are unsure.
1013 Say Y if you want to enable such checks.
1015 config RCU_CPU_STALL_INFO
1016 bool "Print additional diagnostics on RCU CPU stall"
1017 depends on (TREE_RCU || TREE_PREEMPT_RCU) && DEBUG_KERNEL
1020 For each stalled CPU that is aware of the current RCU grace
1021 period, print out additional per-CPU diagnostic information
1022 regarding scheduling-clock ticks, idle state, and,
1023 for RCU_FAST_NO_HZ kernels, idle-entry state.
1025 Say N if you are unsure.
1027 Say Y if you want to enable such diagnostics.
1030 bool "Enable tracing for RCU"
1031 depends on DEBUG_KERNEL
1034 This option provides tracing in RCU which presents stats
1035 in debugfs for debugging RCU implementation.
1037 Say Y here if you want to enable RCU tracing
1038 Say N if you are unsure.
1040 endmenu # "RCU Debugging"
1042 config KPROBES_SANITY_TEST
1043 bool "Kprobes sanity tests"
1044 depends on DEBUG_KERNEL
1048 This option provides for testing basic kprobes functionality on
1049 boot. A sample kprobe, jprobe and kretprobe are inserted and
1050 verified for functionality.
1052 Say N if you are unsure.
1054 config BACKTRACE_SELF_TEST
1055 tristate "Self test for the backtrace code"
1056 depends on DEBUG_KERNEL
1059 This option provides a kernel module that can be used to test
1060 the kernel stack backtrace code. This option is not useful
1061 for distributions or general kernels, but only for kernel
1062 developers working on architecture code.
1064 Note that if you want to also test saved backtraces, you will
1065 have to enable STACKTRACE as well.
1067 Say N if you are unsure.
1069 config DEBUG_BLOCK_EXT_DEVT
1070 bool "Force extended block device numbers and spread them"
1071 depends on DEBUG_KERNEL
1075 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON
1076 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT
1077 YOU ARE DOING. Distros, please enable this and fix whatever
1080 Conventionally, block device numbers are allocated from
1081 predetermined contiguous area. However, extended block area
1082 may introduce non-contiguous block device numbers. This
1083 option forces most block device numbers to be allocated from
1084 the extended space and spreads them to discover kernel or
1085 userland code paths which assume predetermined contiguous
1086 device number allocation.
1088 Note that turning on this debug option shuffles all the
1089 device numbers for all IDE and SCSI devices including libata
1090 ones, so root partition specified using device number
1091 directly (via rdev or root=MAJ:MIN) won't work anymore.
1092 Textual device names (root=/dev/sdXn) will continue to work.
1094 Say N if you are unsure.
1096 config DEBUG_FORCE_WEAK_PER_CPU
1097 bool "Force weak per-cpu definitions"
1098 depends on DEBUG_KERNEL
1100 s390 and alpha require percpu variables in modules to be
1101 defined weak to work around addressing range issue which
1102 puts the following two restrictions on percpu variable
1105 1. percpu symbols must be unique whether static or not
1106 2. percpu variables can't be defined inside a function
1108 To ensure that generic code follows the above rules, this
1109 option forces all percpu variables to be defined as weak.
1111 config DEBUG_PER_CPU_MAPS
1112 bool "Debug access to per_cpu maps"
1113 depends on DEBUG_KERNEL
1116 Say Y to verify that the per_cpu map being accessed has
1117 been set up. This adds a fair amount of code to kernel memory
1118 and decreases performance.
1123 tristate "Linux Kernel Dump Test Tool Module"
1128 This module enables testing of the different dumping mechanisms by
1129 inducing system failures at predefined crash points.
1130 If you don't need it: say N
1131 Choose M here to compile this code as a module. The module will be
1134 Documentation on how to use the module can be found in
1135 Documentation/fault-injection/provoke-crashes.txt
1137 config NOTIFIER_ERROR_INJECTION
1138 tristate "Notifier error injection"
1139 depends on DEBUG_KERNEL
1142 This option provides the ability to inject artificial errors to
1143 specified notifier chain callbacks. It is useful to test the error
1144 handling of notifier call chain failures.
1148 config CPU_NOTIFIER_ERROR_INJECT
1149 tristate "CPU notifier error injection module"
1150 depends on HOTPLUG_CPU && NOTIFIER_ERROR_INJECTION
1152 This option provides a kernel module that can be used to test
1153 the error handling of the cpu notifiers by injecting artificial
1154 errors to CPU notifier chain callbacks. It is controlled through
1155 debugfs interface under /sys/kernel/debug/notifier-error-inject/cpu
1157 If the notifier call chain should be failed with some events
1158 notified, write the error code to "actions/<notifier event>/error".
1160 Example: Inject CPU offline error (-1 == -EPERM)
1162 # cd /sys/kernel/debug/notifier-error-inject/cpu
1163 # echo -1 > actions/CPU_DOWN_PREPARE/error
1164 # echo 0 > /sys/devices/system/cpu/cpu1/online
1165 bash: echo: write error: Operation not permitted
1167 To compile this code as a module, choose M here: the module will
1168 be called cpu-notifier-error-inject.
1172 config PM_NOTIFIER_ERROR_INJECT
1173 tristate "PM notifier error injection module"
1174 depends on PM && NOTIFIER_ERROR_INJECTION
1175 default m if PM_DEBUG
1177 This option provides the ability to inject artificial errors to
1178 PM notifier chain callbacks. It is controlled through debugfs
1179 interface /sys/kernel/debug/notifier-error-inject/pm
1181 If the notifier call chain should be failed with some events
1182 notified, write the error code to "actions/<notifier event>/error".
1184 Example: Inject PM suspend error (-12 = -ENOMEM)
1186 # cd /sys/kernel/debug/notifier-error-inject/pm/
1187 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1188 # echo mem > /sys/power/state
1189 bash: echo: write error: Cannot allocate memory
1191 To compile this code as a module, choose M here: the module will
1192 be called pm-notifier-error-inject.
1196 config MEMORY_NOTIFIER_ERROR_INJECT
1197 tristate "Memory hotplug notifier error injection module"
1198 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION
1200 This option provides the ability to inject artificial errors to
1201 memory hotplug notifier chain callbacks. It is controlled through
1202 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
1204 If the notifier call chain should be failed with some events
1205 notified, write the error code to "actions/<notifier event>/error".
1207 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
1209 # cd /sys/kernel/debug/notifier-error-inject/memory
1210 # echo -12 > actions/MEM_GOING_OFFLINE/error
1211 # echo offline > /sys/devices/system/memory/memoryXXX/state
1212 bash: echo: write error: Cannot allocate memory
1214 To compile this code as a module, choose M here: the module will
1215 be called memory-notifier-error-inject.
1219 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1220 tristate "OF reconfig notifier error injection module"
1221 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1223 This option provides the ability to inject artificial errors to
1224 OF reconfig notifier chain callbacks. It is controlled
1225 through debugfs interface under
1226 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1228 If the notifier call chain should be failed with some events
1229 notified, write the error code to "actions/<notifier event>/error".
1231 To compile this code as a module, choose M here: the module will
1232 be called of-reconfig-notifier-error-inject.
1236 config FAULT_INJECTION
1237 bool "Fault-injection framework"
1238 depends on DEBUG_KERNEL
1240 Provide fault-injection framework.
1241 For more details, see Documentation/fault-injection/.
1244 bool "Fault-injection capability for kmalloc"
1245 depends on FAULT_INJECTION
1246 depends on SLAB || SLUB
1248 Provide fault-injection capability for kmalloc.
1250 config FAIL_PAGE_ALLOC
1251 bool "Fault-injection capabilitiy for alloc_pages()"
1252 depends on FAULT_INJECTION
1254 Provide fault-injection capability for alloc_pages().
1256 config FAIL_MAKE_REQUEST
1257 bool "Fault-injection capability for disk IO"
1258 depends on FAULT_INJECTION && BLOCK
1260 Provide fault-injection capability for disk IO.
1262 config FAIL_IO_TIMEOUT
1263 bool "Fault-injection capability for faking disk interrupts"
1264 depends on FAULT_INJECTION && BLOCK
1266 Provide fault-injection capability on end IO handling. This
1267 will make the block layer "forget" an interrupt as configured,
1268 thus exercising the error handling.
1270 Only works with drivers that use the generic timeout handling,
1271 for others it wont do anything.
1273 config FAIL_MMC_REQUEST
1274 bool "Fault-injection capability for MMC IO"
1276 depends on FAULT_INJECTION && MMC
1278 Provide fault-injection capability for MMC IO.
1279 This will make the mmc core return data errors. This is
1280 useful to test the error handling in the mmc block device
1281 and to test how the mmc host driver handles retries from
1284 config FAULT_INJECTION_DEBUG_FS
1285 bool "Debugfs entries for fault-injection capabilities"
1286 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1288 Enable configuration of fault-injection capabilities via debugfs.
1290 config FAULT_INJECTION_STACKTRACE_FILTER
1291 bool "stacktrace filter for fault-injection capabilities"
1292 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1295 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND
1297 Provide stacktrace filter for fault-injection capabilities
1300 bool "Latency measuring infrastructure"
1301 depends on HAVE_LATENCYTOP_SUPPORT
1302 depends on DEBUG_KERNEL
1303 depends on STACKTRACE_SUPPORT
1305 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND
1312 Enable this option if you want to use the LatencyTOP tool
1313 to find out which userspace is blocking on what kernel operations.
1315 config ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS
1318 config DEBUG_STRICT_USER_COPY_CHECKS
1319 bool "Strict user copy size checks"
1320 depends on ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS
1321 depends on DEBUG_KERNEL && !TRACE_BRANCH_PROFILING
1323 Enabling this option turns a certain set of sanity checks for user
1324 copy operations into compile time failures.
1326 The copy_from_user() etc checks are there to help test if there
1327 are sufficient security checks on the length argument of
1328 the copy operation, by having gcc prove that the argument is
1333 source mm/Kconfig.debug
1334 source kernel/trace/Kconfig
1337 tristate "Red-Black tree test"
1338 depends on m && DEBUG_KERNEL
1340 A benchmark measuring the performance of the rbtree library.
1341 Also includes rbtree invariant checks.
1343 config INTERVAL_TREE_TEST
1344 tristate "Interval tree test"
1345 depends on m && DEBUG_KERNEL
1347 A benchmark measuring the performance of the interval tree library
1349 config PROVIDE_OHCI1394_DMA_INIT
1350 bool "Remote debugging over FireWire early on boot"
1351 depends on PCI && X86
1353 If you want to debug problems which hang or crash the kernel early
1354 on boot and the crashing machine has a FireWire port, you can use
1355 this feature to remotely access the memory of the crashed machine
1356 over FireWire. This employs remote DMA as part of the OHCI1394
1357 specification which is now the standard for FireWire controllers.
1359 With remote DMA, you can monitor the printk buffer remotely using
1360 firescope and access all memory below 4GB using fireproxy from gdb.
1361 Even controlling a kernel debugger is possible using remote DMA.
1365 If ohci1394_dma=early is used as boot parameter, it will initialize
1366 all OHCI1394 controllers which are found in the PCI config space.
1368 As all changes to the FireWire bus such as enabling and disabling
1369 devices cause a bus reset and thereby disable remote DMA for all
1370 devices, be sure to have the cable plugged and FireWire enabled on
1371 the debugging host before booting the debug target for debugging.
1373 This code (~1k) is freed after boot. By then, the firewire stack
1374 in charge of the OHCI-1394 controllers should be used instead.
1376 See Documentation/debugging-via-ohci1394.txt for more information.
1378 config FIREWIRE_OHCI_REMOTE_DMA
1379 bool "Remote debugging over FireWire with firewire-ohci"
1380 depends on FIREWIRE_OHCI
1382 This option lets you use the FireWire bus for remote debugging
1383 with help of the firewire-ohci driver. It enables unfiltered
1384 remote DMA in firewire-ohci.
1385 See Documentation/debugging-via-ohci1394.txt for more information.
1390 bool "Build targets in Documentation/ tree"
1391 depends on HEADERS_CHECK
1393 This option attempts to build objects from the source files in the
1394 kernel Documentation/ tree.
1396 Say N if you are unsure.
1398 config DYNAMIC_DEBUG
1399 bool "Enable dynamic printk() support"
1405 Compiles debug level messages into the kernel, which would not
1406 otherwise be available at runtime. These messages can then be
1407 enabled/disabled based on various levels of scope - per source file,
1408 function, module, format string, and line number. This mechanism
1409 implicitly compiles in all pr_debug() and dev_dbg() calls, which
1410 enlarges the kernel text size by about 2%.
1412 If a source file is compiled with DEBUG flag set, any
1413 pr_debug() calls in it are enabled by default, but can be
1414 disabled at runtime as below. Note that DEBUG flag is
1415 turned on by many CONFIG_*DEBUG* options.
1419 Dynamic debugging is controlled via the 'dynamic_debug/control' file,
1420 which is contained in the 'debugfs' filesystem. Thus, the debugfs
1421 filesystem must first be mounted before making use of this feature.
1422 We refer the control file as: <debugfs>/dynamic_debug/control. This
1423 file contains a list of the debug statements that can be enabled. The
1424 format for each line of the file is:
1426 filename:lineno [module]function flags format
1428 filename : source file of the debug statement
1429 lineno : line number of the debug statement
1430 module : module that contains the debug statement
1431 function : function that contains the debug statement
1432 flags : '=p' means the line is turned 'on' for printing
1433 format : the format used for the debug statement
1437 nullarbor:~ # cat <debugfs>/dynamic_debug/control
1438 # filename:lineno [module]function flags format
1439 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012"
1440 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012"
1441 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012"
1445 // enable the message at line 1603 of file svcsock.c
1446 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' >
1447 <debugfs>/dynamic_debug/control
1449 // enable all the messages in file svcsock.c
1450 nullarbor:~ # echo -n 'file svcsock.c +p' >
1451 <debugfs>/dynamic_debug/control
1453 // enable all the messages in the NFS server module
1454 nullarbor:~ # echo -n 'module nfsd +p' >
1455 <debugfs>/dynamic_debug/control
1457 // enable all 12 messages in the function svc_process()
1458 nullarbor:~ # echo -n 'func svc_process +p' >
1459 <debugfs>/dynamic_debug/control
1461 // disable all 12 messages in the function svc_process()
1462 nullarbor:~ # echo -n 'func svc_process -p' >
1463 <debugfs>/dynamic_debug/control
1465 See Documentation/dynamic-debug-howto.txt for additional information.
1467 config DMA_API_DEBUG
1468 bool "Enable debugging of DMA-API usage"
1469 depends on HAVE_DMA_API_DEBUG
1471 Enable this option to debug the use of the DMA API by device drivers.
1472 With this option you will be able to detect common bugs in device
1473 drivers like double-freeing of DMA mappings or freeing mappings that
1474 were never allocated.
1475 This option causes a performance degredation. Use only if you want
1476 to debug device drivers. If unsure, say N.
1478 config ATOMIC64_SELFTEST
1479 bool "Perform an atomic64_t self-test at boot"
1481 Enable this option to test the atomic64_t functions at boot.
1485 config ASYNC_RAID6_TEST
1486 tristate "Self test for hardware accelerated raid6 recovery"
1487 depends on ASYNC_RAID6_RECOV
1490 This is a one-shot self test that permutes through the
1491 recovery of all the possible two disk failure scenarios for a
1492 N-disk array. Recovery is performed with the asynchronous
1493 raid6 recovery routines, and will optionally use an offload
1494 engine if one is available.
1498 source "samples/Kconfig"
1500 source "lib/Kconfig.kgdb"
1502 source "lib/Kconfig.kmemcheck"
1504 config TEST_STRING_HELPERS
1505 tristate "Test functions located in the string_helpers module at runtime"
1508 tristate "Test kstrto*() family of functions at runtime"