3 bool "Show timing information on printks"
6 Selecting this option causes timing information to be
7 included in printk output. This allows you to measure
8 the interval between kernel operations, including bootup
9 operations. This is useful for identifying long delays
12 config ENABLE_WARN_DEPRECATED
13 bool "Enable __deprecated logic"
16 Enable the __deprecated logic in the kernel build.
17 Disable this to suppress the "warning: 'foo' is deprecated
18 (declared at kernel/power/somefile.c:1234)" messages.
20 config ENABLE_MUST_CHECK
21 bool "Enable __must_check logic"
24 Enable the __must_check logic in the kernel build. Disable this to
25 suppress the "warning: ignoring return value of 'foo', declared with
26 attribute warn_unused_result" messages.
29 bool "Magic SysRq key"
32 If you say Y here, you will have some control over the system even
33 if the system crashes for example during kernel debugging (e.g., you
34 will be able to flush the buffer cache to disk, reboot the system
35 immediately or dump some status information). This is accomplished
36 by pressing various keys while holding SysRq (Alt+PrintScreen). It
37 also works on a serial console (on PC hardware at least), if you
38 send a BREAK and then within 5 seconds a command keypress. The
39 keys are documented in <file:Documentation/sysrq.txt>. Don't say Y
40 unless you really know what this hack does.
43 bool "Enable unused/obsolete exported symbols"
46 Unused but exported symbols make the kernel needlessly bigger. For
47 that reason most of these unused exports will soon be removed. This
48 option is provided temporarily to provide a transition period in case
49 some external kernel module needs one of these symbols anyway. If you
50 encounter such a case in your module, consider if you are actually
51 using the right API. (rationale: since nobody in the kernel is using
52 this in a module, there is a pretty good chance it's actually the
53 wrong interface to use). If you really need the symbol, please send a
54 mail to the linux kernel mailing list mentioning the symbol and why
55 you really need it, and what the merge plan to the mainline kernel for
59 bool "Debug Filesystem"
62 debugfs is a virtual file system that kernel developers use to put
63 debugging files into. Enable this option to be able to read and
69 bool "Run 'make headers_check' when building vmlinux"
72 This option will extract the user-visible kernel headers whenever
73 building the kernel, and will run basic sanity checks on them to
74 ensure that exported files do not attempt to include files which
75 were not exported, etc.
77 If you're making modifications to header files which are
78 relevant for userspace, say 'Y', and check the headers
79 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in
80 your build tree), to make sure they're suitable.
82 config DEBUG_SECTION_MISMATCH
83 bool "Enable full Section mismatch analysis"
85 # This option is on purpose disabled for now.
86 # It will be enabled when we are down to a resonable number
87 # of section mismatch warnings (< 10 for an allyesconfig build)
89 The section mismatch analysis checks if there are illegal
90 references from one section to another section.
91 Linux will during link or during runtime drop some sections
92 and any use of code/data previously in these sections will
93 most likely result in an oops.
94 In the code functions and variables are annotated with
95 __init, __devinit etc. (see full list in include/linux/init.h)
96 which results in the code/data being placed in specific sections.
97 The section mismatch analysis is always done after a full
98 kernel build but enabling this option will in addition
100 - Add the option -fno-inline-functions-called-once to gcc
101 When inlining a function annotated __init in a non-init
102 function we would lose the section information and thus
103 the analysis would not catch the illegal reference.
104 This option tells gcc to inline less but will also
105 result in a larger kernel.
106 - Run the section mismatch analysis for each module/built-in.o
107 When we run the section mismatch analysis on vmlinux.o we
108 lose valueble information about where the mismatch was
110 Running the analysis for each module/built-in.o file
111 will tell where the mismatch happens much closer to the
112 source. The drawback is that we will report the same
113 mismatch at least twice.
114 - Enable verbose reporting from modpost to help solving
115 the section mismatches reported.
118 bool "Kernel debugging"
120 Say Y here if you are developing drivers or trying to debug and
121 identify kernel problems.
124 bool "Debug shared IRQ handlers"
125 depends on DEBUG_KERNEL && GENERIC_HARDIRQS
127 Enable this to generate a spurious interrupt as soon as a shared
128 interrupt handler is registered, and just before one is deregistered.
129 Drivers ought to be able to handle interrupts coming in at those
130 points; some don't and need to be caught.
132 config DETECT_SOFTLOCKUP
133 bool "Detect Soft Lockups"
134 depends on DEBUG_KERNEL && !S390
137 Say Y here to enable the kernel to detect "soft lockups",
138 which are bugs that cause the kernel to loop in kernel
139 mode for more than 10 seconds, without giving other tasks a
142 When a soft-lockup is detected, the kernel will print the
143 current stack trace (which you should report), but the
144 system will stay locked up. This feature has negligible
147 (Note that "hard lockups" are separate type of bugs that
148 can be detected via the NMI-watchdog, on platforms that
152 bool "Collect scheduler debugging info"
153 depends on DEBUG_KERNEL && PROC_FS
156 If you say Y here, the /proc/sched_debug file will be provided
157 that can help debug the scheduler. The runtime overhead of this
161 bool "Collect scheduler statistics"
162 depends on DEBUG_KERNEL && PROC_FS
164 If you say Y here, additional code will be inserted into the
165 scheduler and related routines to collect statistics about
166 scheduler behavior and provide them in /proc/schedstat. These
167 stats may be useful for both tuning and debugging the scheduler
168 If you aren't debugging the scheduler or trying to tune a specific
169 application, you can say N to avoid the very slight overhead
173 bool "Collect kernel timers statistics"
174 depends on DEBUG_KERNEL && PROC_FS
176 If you say Y here, additional code will be inserted into the
177 timer routines to collect statistics about kernel timers being
178 reprogrammed. The statistics can be read from /proc/timer_stats.
179 The statistics collection is started by writing 1 to /proc/timer_stats,
180 writing 0 stops it. This feature is useful to collect information
181 about timer usage patterns in kernel and userspace. This feature
182 is lightweight if enabled in the kernel config but not activated
183 (it defaults to deactivated on bootup and will only be activated
184 if some application like powertop activates it explicitly).
187 bool "Debug slab memory allocations"
188 depends on DEBUG_KERNEL && SLAB
190 Say Y here to have the kernel do limited verification on memory
191 allocation as well as poisoning memory on free to catch use of freed
192 memory. This can make kmalloc/kfree-intensive workloads much slower.
194 config DEBUG_SLAB_LEAK
195 bool "Memory leak debugging"
196 depends on DEBUG_SLAB
199 bool "SLUB debugging on by default"
200 depends on SLUB && SLUB_DEBUG
203 Boot with debugging on by default. SLUB boots by default with
204 the runtime debug capabilities switched off. Enabling this is
205 equivalent to specifying the "slub_debug" parameter on boot.
206 There is no support for more fine grained debug control like
207 possible with slub_debug=xxx. SLUB debugging may be switched
208 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
213 bool "Enable SLUB performance statistics"
216 SLUB statistics are useful to debug SLUBs allocation behavior in
217 order find ways to optimize the allocator. This should never be
218 enabled for production use since keeping statistics slows down
219 the allocator by a few percentage points. The slabinfo command
220 supports the determination of the most active slabs to figure
221 out which slabs are relevant to a particular load.
222 Try running: slabinfo -DA
225 bool "Debug preemptible kernel"
226 depends on DEBUG_KERNEL && PREEMPT && (TRACE_IRQFLAGS_SUPPORT || PPC64)
229 If you say Y here then the kernel will use a debug variant of the
230 commonly used smp_processor_id() function and will print warnings
231 if kernel code uses it in a preemption-unsafe way. Also, the kernel
232 will detect preemption count underflows.
234 config DEBUG_RT_MUTEXES
235 bool "RT Mutex debugging, deadlock detection"
236 depends on DEBUG_KERNEL && RT_MUTEXES
238 This allows rt mutex semantics violations and rt mutex related
239 deadlocks (lockups) to be detected and reported automatically.
244 depends on DEBUG_RT_MUTEXES
246 config RT_MUTEX_TESTER
247 bool "Built-in scriptable tester for rt-mutexes"
248 depends on DEBUG_KERNEL && RT_MUTEXES
250 This option enables a rt-mutex tester.
252 config DEBUG_SPINLOCK
253 bool "Spinlock and rw-lock debugging: basic checks"
254 depends on DEBUG_KERNEL
256 Say Y here and build SMP to catch missing spinlock initialization
257 and certain other kinds of spinlock errors commonly made. This is
258 best used in conjunction with the NMI watchdog so that spinlock
259 deadlocks are also debuggable.
262 bool "Mutex debugging: basic checks"
263 depends on DEBUG_KERNEL
265 This feature allows mutex semantics violations to be detected and
268 config DEBUG_SEMAPHORE
269 bool "Semaphore debugging"
270 depends on DEBUG_KERNEL
271 depends on ALPHA || FRV
274 If you say Y here then semaphore processing will issue lots of
275 verbose debugging messages. If you suspect a semaphore problem or a
276 kernel hacker asks for this option then say Y. Otherwise say N.
278 config DEBUG_LOCK_ALLOC
279 bool "Lock debugging: detect incorrect freeing of live locks"
280 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
281 select DEBUG_SPINLOCK
285 This feature will check whether any held lock (spinlock, rwlock,
286 mutex or rwsem) is incorrectly freed by the kernel, via any of the
287 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
288 vfree(), etc.), whether a live lock is incorrectly reinitialized via
289 spin_lock_init()/mutex_init()/etc., or whether there is any lock
290 held during task exit.
293 bool "Lock debugging: prove locking correctness"
294 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
296 select DEBUG_SPINLOCK
298 select DEBUG_LOCK_ALLOC
301 This feature enables the kernel to prove that all locking
302 that occurs in the kernel runtime is mathematically
303 correct: that under no circumstance could an arbitrary (and
304 not yet triggered) combination of observed locking
305 sequences (on an arbitrary number of CPUs, running an
306 arbitrary number of tasks and interrupt contexts) cause a
309 In short, this feature enables the kernel to report locking
310 related deadlocks before they actually occur.
312 The proof does not depend on how hard and complex a
313 deadlock scenario would be to trigger: how many
314 participant CPUs, tasks and irq-contexts would be needed
315 for it to trigger. The proof also does not depend on
316 timing: if a race and a resulting deadlock is possible
317 theoretically (no matter how unlikely the race scenario
318 is), it will be proven so and will immediately be
319 reported by the kernel (once the event is observed that
320 makes the deadlock theoretically possible).
322 If a deadlock is impossible (i.e. the locking rules, as
323 observed by the kernel, are mathematically correct), the
324 kernel reports nothing.
326 NOTE: this feature can also be enabled for rwlocks, mutexes
327 and rwsems - in which case all dependencies between these
328 different locking variants are observed and mapped too, and
329 the proof of observed correctness is also maintained for an
330 arbitrary combination of these separate locking variants.
332 For more details, see Documentation/lockdep-design.txt.
336 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
338 select FRAME_POINTER if !X86 && !MIPS
343 bool "Lock usage statistics"
344 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
346 select DEBUG_SPINLOCK
348 select DEBUG_LOCK_ALLOC
351 This feature enables tracking lock contention points
353 For more details, see Documentation/lockstat.txt
356 bool "Lock dependency engine debugging"
357 depends on DEBUG_KERNEL && LOCKDEP
359 If you say Y here, the lock dependency engine will do
360 additional runtime checks to debug itself, at the price
361 of more runtime overhead.
363 config TRACE_IRQFLAGS
364 depends on DEBUG_KERNEL
367 depends on TRACE_IRQFLAGS_SUPPORT
368 depends on PROVE_LOCKING
370 config DEBUG_SPINLOCK_SLEEP
371 bool "Spinlock debugging: sleep-inside-spinlock checking"
372 depends on DEBUG_KERNEL
374 If you say Y here, various routines which may sleep will become very
375 noisy if they are called with a spinlock held.
377 config DEBUG_LOCKING_API_SELFTESTS
378 bool "Locking API boot-time self-tests"
379 depends on DEBUG_KERNEL
381 Say Y here if you want the kernel to run a short self-test during
382 bootup. The self-test checks whether common types of locking bugs
383 are detected by debugging mechanisms or not. (if you disable
384 lock debugging then those bugs wont be detected of course.)
385 The following locking APIs are covered: spinlocks, rwlocks,
390 depends on DEBUG_KERNEL
391 depends on STACKTRACE_SUPPORT
394 bool "kobject debugging"
395 depends on DEBUG_KERNEL
397 If you say Y here, some extra kobject debugging messages will be sent
401 bool "Highmem debugging"
402 depends on DEBUG_KERNEL && HIGHMEM
404 This options enables addition error checking for high memory systems.
405 Disable for production systems.
407 config DEBUG_BUGVERBOSE
408 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED
410 depends on ARM || AVR32 || M32R || M68K || SPARC32 || SPARC64 || \
411 FRV || SUPERH || GENERIC_BUG || BLACKFIN || MN10300
414 Say Y here to make BUG() panics output the file name and line number
415 of the BUG call as well as the EIP and oops trace. This aids
416 debugging but costs about 70-100K of memory.
419 bool "Compile the kernel with debug info"
420 depends on DEBUG_KERNEL
422 If you say Y here the resulting kernel image will include
423 debugging info resulting in a larger kernel image.
424 This adds debug symbols to the kernel and modules (gcc -g), and
425 is needed if you intend to use kernel crashdump or binary object
426 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
427 Say Y here only if you plan to debug the kernel.
433 depends on DEBUG_KERNEL
435 Enable this to turn on extended checks in the virtual-memory system
436 that may impact performance.
441 bool "Debug linked list manipulation"
442 depends on DEBUG_KERNEL
444 Enable this to turn on extended checks in the linked-list
450 bool "Debug SG table operations"
451 depends on DEBUG_KERNEL
453 Enable this to turn on checks on scatter-gather tables. This can
454 help find problems with drivers that do not properly initialize
460 bool "Compile the kernel with frame pointers"
461 depends on DEBUG_KERNEL && \
462 (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390 || \
463 AVR32 || SUPERH || BLACKFIN || MN10300)
464 default y if DEBUG_INFO && UML
466 If you say Y here the resulting kernel image will be slightly larger
467 and slower, but it might give very useful debugging information on
468 some architectures or if you use external debuggers.
469 If you don't debug the kernel, you can say N.
471 config BOOT_PRINTK_DELAY
472 bool "Delay each boot printk message by N milliseconds"
473 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY
475 This build option allows you to read kernel boot messages
476 by inserting a short delay after each one. The delay is
477 specified in milliseconds on the kernel command line,
478 using "boot_delay=N".
480 It is likely that you would also need to use "lpj=M" to preset
481 the "loops per jiffie" value.
482 See a previous boot log for the "lpj" value to use for your
483 system, and then set "lpj=M" before setting "boot_delay=N".
484 NOTE: Using this option may adversely affect SMP systems.
485 I.e., processors other than the first one may not boot up.
486 BOOT_PRINTK_DELAY also may cause DETECT_SOFTLOCKUP to detect
487 what it believes to be lockup conditions.
489 config RCU_TORTURE_TEST
490 tristate "torture tests for RCU"
491 depends on DEBUG_KERNEL
495 This option provides a kernel module that runs torture tests
496 on the RCU infrastructure. The kernel module may be built
497 after the fact on the running kernel to be tested, if desired.
499 Say M if you want the RCU torture tests to build as a module.
500 Say N if you are unsure.
502 config KPROBES_SANITY_TEST
503 bool "Kprobes sanity tests"
504 depends on DEBUG_KERNEL
508 This option provides for testing basic kprobes functionality on
509 boot. A sample kprobe, jprobe and kretprobe are inserted and
510 verified for functionality.
512 Say N if you are unsure.
514 config BACKTRACE_SELF_TEST
515 tristate "Self test for the backtrace code"
516 depends on DEBUG_KERNEL
519 This option provides a kernel module that can be used to test
520 the kernel stack backtrace code. This option is not useful
521 for distributions or general kernels, but only for kernel
522 developers working on architecture code.
524 Say N if you are unsure.
527 tristate "Linux Kernel Dump Test Tool Module"
528 depends on DEBUG_KERNEL
533 This module enables testing of the different dumping mechanisms by
534 inducing system failures at predefined crash points.
535 If you don't need it: say N
536 Choose M here to compile this code as a module. The module will be
539 Documentation on how to use the module can be found in
542 config FAULT_INJECTION
543 bool "Fault-injection framework"
544 depends on DEBUG_KERNEL
546 Provide fault-injection framework.
547 For more details, see Documentation/fault-injection/.
550 bool "Fault-injection capability for kmalloc"
551 depends on FAULT_INJECTION
553 Provide fault-injection capability for kmalloc.
555 config FAIL_PAGE_ALLOC
556 bool "Fault-injection capabilitiy for alloc_pages()"
557 depends on FAULT_INJECTION
559 Provide fault-injection capability for alloc_pages().
561 config FAIL_MAKE_REQUEST
562 bool "Fault-injection capability for disk IO"
563 depends on FAULT_INJECTION
565 Provide fault-injection capability for disk IO.
567 config FAULT_INJECTION_DEBUG_FS
568 bool "Debugfs entries for fault-injection capabilities"
569 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
571 Enable configuration of fault-injection capabilities via debugfs.
573 config FAULT_INJECTION_STACKTRACE_FILTER
574 bool "stacktrace filter for fault-injection capabilities"
575 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
580 Provide stacktrace filter for fault-injection capabilities
583 bool "Latency measuring infrastructure"
584 select FRAME_POINTER if !MIPS
590 depends on HAVE_LATENCYTOP_SUPPORT
592 Enable this option if you want to use the LatencyTOP tool
593 to find out which userspace is blocking on what kernel operations.
595 config PROVIDE_OHCI1394_DMA_INIT
596 bool "Provide code for enabling DMA over FireWire early on boot"
597 depends on PCI && X86
599 If you want to debug problems which hang or crash the kernel early
600 on boot and the crashing machine has a FireWire port, you can use
601 this feature to remotely access the memory of the crashed machine
602 over FireWire. This employs remote DMA as part of the OHCI1394
603 specification which is now the standard for FireWire controllers.
605 With remote DMA, you can monitor the printk buffer remotely using
606 firescope and access all memory below 4GB using fireproxy from gdb.
607 Even controlling a kernel debugger is possible using remote DMA.
611 If ohci1394_dma=early is used as boot parameter, it will initialize
612 all OHCI1394 controllers which are found in the PCI config space.
614 As all changes to the FireWire bus such as enabling and disabling
615 devices cause a bus reset and thereby disable remote DMA for all
616 devices, be sure to have the cable plugged and FireWire enabled on
617 the debugging host before booting the debug target for debugging.
619 This code (~1k) is freed after boot. By then, the firewire stack
620 in charge of the OHCI-1394 controllers should be used instead.
622 See Documentation/debugging-via-ohci1394.txt for more information.
624 source "samples/Kconfig"