1 # SPDX-License-Identifier: GPL-2.0-only
4 menu "printk and dmesg options"
7 bool "Show timing information on printks"
10 Selecting this option causes time stamps of the printk()
11 messages to be added to the output of the syslog() system
12 call and at the console.
14 The timestamp is always recorded internally, and exported
15 to /dev/kmsg. This flag just specifies if the timestamp should
16 be included, not that the timestamp is recorded.
18 The behavior is also controlled by the kernel command line
19 parameter printk.time=1. See Documentation/admin-guide/kernel-parameters.rst
22 bool "Show caller information on printks"
25 Selecting this option causes printk() to add a caller "thread id" (if
26 in task context) or a caller "processor id" (if not in task context)
29 This option is intended for environments where multiple threads
30 concurrently call printk() for many times, for it is difficult to
31 interpret without knowing where these lines (or sometimes individual
32 line which was divided into multiple lines due to race) came from.
34 Since toggling after boot makes the code racy, currently there is
35 no option to enable/disable at the kernel command line parameter or
38 config CONSOLE_LOGLEVEL_DEFAULT
39 int "Default console loglevel (1-15)"
43 Default loglevel to determine what will be printed on the console.
45 Setting a default here is equivalent to passing in loglevel=<x> in
46 the kernel bootargs. loglevel=<x> continues to override whatever
47 value is specified here as well.
49 Note: This does not affect the log level of un-prefixed printk()
50 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT
53 config CONSOLE_LOGLEVEL_QUIET
54 int "quiet console loglevel (1-15)"
58 loglevel to use when "quiet" is passed on the kernel commandline.
60 When "quiet" is passed on the kernel commandline this loglevel
61 will be used as the loglevel. IOW passing "quiet" will be the
62 equivalent of passing "loglevel=<CONSOLE_LOGLEVEL_QUIET>"
64 config MESSAGE_LOGLEVEL_DEFAULT
65 int "Default message log level (1-7)"
69 Default log level for printk statements with no specified priority.
71 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks
72 that are auditing their logs closely may want to set it to a lower
75 Note: This does not affect what message level gets printed on the console
76 by default. To change that, use loglevel=<x> in the kernel bootargs,
77 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value.
79 config BOOT_PRINTK_DELAY
80 bool "Delay each boot printk message by N milliseconds"
81 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY
83 This build option allows you to read kernel boot messages
84 by inserting a short delay after each one. The delay is
85 specified in milliseconds on the kernel command line,
88 It is likely that you would also need to use "lpj=M" to preset
89 the "loops per jiffie" value.
90 See a previous boot log for the "lpj" value to use for your
91 system, and then set "lpj=M" before setting "boot_delay=N".
92 NOTE: Using this option may adversely affect SMP systems.
93 I.e., processors other than the first one may not boot up.
94 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect
95 what it believes to be lockup conditions.
98 bool "Enable dynamic printk() support"
104 Compiles debug level messages into the kernel, which would not
105 otherwise be available at runtime. These messages can then be
106 enabled/disabled based on various levels of scope - per source file,
107 function, module, format string, and line number. This mechanism
108 implicitly compiles in all pr_debug() and dev_dbg() calls, which
109 enlarges the kernel text size by about 2%.
111 If a source file is compiled with DEBUG flag set, any
112 pr_debug() calls in it are enabled by default, but can be
113 disabled at runtime as below. Note that DEBUG flag is
114 turned on by many CONFIG_*DEBUG* options.
118 Dynamic debugging is controlled via the 'dynamic_debug/control' file,
119 which is contained in the 'debugfs' filesystem. Thus, the debugfs
120 filesystem must first be mounted before making use of this feature.
121 We refer the control file as: <debugfs>/dynamic_debug/control. This
122 file contains a list of the debug statements that can be enabled. The
123 format for each line of the file is:
125 filename:lineno [module]function flags format
127 filename : source file of the debug statement
128 lineno : line number of the debug statement
129 module : module that contains the debug statement
130 function : function that contains the debug statement
131 flags : '=p' means the line is turned 'on' for printing
132 format : the format used for the debug statement
136 nullarbor:~ # cat <debugfs>/dynamic_debug/control
137 # filename:lineno [module]function flags format
138 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012"
139 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012"
140 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012"
144 // enable the message at line 1603 of file svcsock.c
145 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' >
146 <debugfs>/dynamic_debug/control
148 // enable all the messages in file svcsock.c
149 nullarbor:~ # echo -n 'file svcsock.c +p' >
150 <debugfs>/dynamic_debug/control
152 // enable all the messages in the NFS server module
153 nullarbor:~ # echo -n 'module nfsd +p' >
154 <debugfs>/dynamic_debug/control
156 // enable all 12 messages in the function svc_process()
157 nullarbor:~ # echo -n 'func svc_process +p' >
158 <debugfs>/dynamic_debug/control
160 // disable all 12 messages in the function svc_process()
161 nullarbor:~ # echo -n 'func svc_process -p' >
162 <debugfs>/dynamic_debug/control
164 See Documentation/admin-guide/dynamic-debug-howto.rst for additional
167 endmenu # "printk and dmesg options"
169 menu "Compile-time checks and compiler options"
172 bool "Compile the kernel with debug info"
173 depends on DEBUG_KERNEL && !COMPILE_TEST
175 If you say Y here the resulting kernel image will include
176 debugging info resulting in a larger kernel image.
177 This adds debug symbols to the kernel and modules (gcc -g), and
178 is needed if you intend to use kernel crashdump or binary object
179 tools like crash, kgdb, LKCD, gdb, etc on the kernel.
180 Say Y here only if you plan to debug the kernel.
184 config DEBUG_INFO_REDUCED
185 bool "Reduce debugging information"
186 depends on DEBUG_INFO
188 If you say Y here gcc is instructed to generate less debugging
189 information for structure types. This means that tools that
190 need full debugging information (like kgdb or systemtap) won't
191 be happy. But if you merely need debugging information to
192 resolve line numbers there is no loss. Advantage is that
193 build directory object sizes shrink dramatically over a full
194 DEBUG_INFO build and compile times are reduced too.
195 Only works with newer gcc versions.
197 config DEBUG_INFO_SPLIT
198 bool "Produce split debuginfo in .dwo files"
199 depends on DEBUG_INFO
200 depends on $(cc-option,-gsplit-dwarf)
202 Generate debug info into separate .dwo files. This significantly
203 reduces the build directory size for builds with DEBUG_INFO,
204 because it stores the information only once on disk in .dwo
205 files instead of multiple times in object files and executables.
206 In addition the debug information is also compressed.
208 Requires recent gcc (4.7+) and recent gdb/binutils.
209 Any tool that packages or reads debug information would need
210 to know about the .dwo files and include them.
211 Incompatible with older versions of ccache.
213 config DEBUG_INFO_DWARF4
214 bool "Generate dwarf4 debuginfo"
215 depends on DEBUG_INFO
216 depends on $(cc-option,-gdwarf-4)
218 Generate dwarf4 debug info. This requires recent versions
219 of gcc and gdb. It makes the debug information larger.
220 But it significantly improves the success of resolving
221 variables in gdb on optimized code.
223 config DEBUG_INFO_BTF
224 bool "Generate BTF typeinfo"
225 depends on DEBUG_INFO
227 Generate deduplicated BTF type information from DWARF debug info.
228 Turning this on expects presence of pahole tool, which will convert
229 DWARF type info into equivalent deduplicated BTF type info.
232 bool "Provide GDB scripts for kernel debugging"
233 depends on DEBUG_INFO
235 This creates the required links to GDB helper scripts in the
236 build directory. If you load vmlinux into gdb, the helper
237 scripts will be automatically imported by gdb as well, and
238 additional functions are available to analyze a Linux kernel
239 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst
242 config ENABLE_MUST_CHECK
243 bool "Enable __must_check logic"
246 Enable the __must_check logic in the kernel build. Disable this to
247 suppress the "warning: ignoring return value of 'foo', declared with
248 attribute warn_unused_result" messages.
251 int "Warn for stack frames larger than (needs gcc 4.4)"
253 default 2048 if GCC_PLUGIN_LATENT_ENTROPY
254 default 1280 if (!64BIT && PARISC)
255 default 1024 if (!64BIT && !PARISC)
256 default 2048 if 64BIT
258 Tell gcc to warn at build time for stack frames larger than this.
259 Setting this too low will cause a lot of warnings.
260 Setting it to 0 disables the warning.
263 config STRIP_ASM_SYMS
264 bool "Strip assembler-generated symbols during link"
267 Strip internal assembler-generated symbols during a link (symbols
268 that look like '.Lxxx') so they don't pollute the output of
269 get_wchan() and suchlike.
272 bool "Generate readable assembler code"
273 depends on DEBUG_KERNEL
275 Disable some compiler optimizations that tend to generate human unreadable
276 assembler output. This may make the kernel slightly slower, but it helps
277 to keep kernel developers who have to stare a lot at assembler listings
280 config UNUSED_SYMBOLS
281 bool "Enable unused/obsolete exported symbols"
284 Unused but exported symbols make the kernel needlessly bigger. For
285 that reason most of these unused exports will soon be removed. This
286 option is provided temporarily to provide a transition period in case
287 some external kernel module needs one of these symbols anyway. If you
288 encounter such a case in your module, consider if you are actually
289 using the right API. (rationale: since nobody in the kernel is using
290 this in a module, there is a pretty good chance it's actually the
291 wrong interface to use). If you really need the symbol, please send a
292 mail to the linux kernel mailing list mentioning the symbol and why
293 you really need it, and what the merge plan to the mainline kernel for
297 bool "Debug Filesystem"
299 debugfs is a virtual file system that kernel developers use to put
300 debugging files into. Enable this option to be able to read and
301 write to these files.
303 For detailed documentation on the debugfs API, see
304 Documentation/filesystems/.
308 config HEADERS_INSTALL
309 bool "Install uapi headers to usr/include"
312 This option will install uapi headers (headers exported to user-space)
313 into the usr/include directory for use during the kernel build.
314 This is unneeded for building the kernel itself, but needed for some
315 user-space program samples. It is also needed by some features such
316 as uapi header sanity checks.
319 bool "Run sanity checks on uapi headers when building 'all'"
320 depends on HEADERS_INSTALL
322 This option will run basic sanity checks on uapi headers when
323 building the 'all' target, for example, ensure that they do not
324 attempt to include files which were not exported, etc.
326 If you're making modifications to header files which are
327 relevant for userspace, say 'Y'.
329 config OPTIMIZE_INLINING
330 bool "Allow compiler to uninline functions marked 'inline'"
332 This option determines if the kernel forces gcc to inline the functions
333 developers have marked 'inline'. Doing so takes away freedom from gcc to
334 do what it thinks is best, which is desirable for the gcc 3.x series of
335 compilers. The gcc 4.x series have a rewritten inlining algorithm and
336 enabling this option will generate a smaller kernel there. Hopefully
337 this algorithm is so good that allowing gcc 4.x and above to make the
338 decision will become the default in the future. Until then this option
339 is there to test gcc for this.
343 config DEBUG_SECTION_MISMATCH
344 bool "Enable full Section mismatch analysis"
346 The section mismatch analysis checks if there are illegal
347 references from one section to another section.
348 During linktime or runtime, some sections are dropped;
349 any use of code/data previously in these sections would
350 most likely result in an oops.
351 In the code, functions and variables are annotated with
352 __init,, etc. (see the full list in include/linux/init.h),
353 which results in the code/data being placed in specific sections.
354 The section mismatch analysis is always performed after a full
355 kernel build, and enabling this option causes the following
356 additional step to occur:
357 - Add the option -fno-inline-functions-called-once to gcc commands.
358 When inlining a function annotated with __init in a non-init
359 function, we would lose the section information and thus
360 the analysis would not catch the illegal reference.
361 This option tells gcc to inline less (but it does result in
364 config SECTION_MISMATCH_WARN_ONLY
365 bool "Make section mismatch errors non-fatal"
368 If you say N here, the build process will fail if there are any
369 section mismatch, instead of just throwing warnings.
374 # Select this config option from the architecture Kconfig, if it
375 # is preferred to always offer frame pointers as a config
376 # option on the architecture (regardless of KERNEL_DEBUG):
378 config ARCH_WANT_FRAME_POINTERS
382 bool "Compile the kernel with frame pointers"
383 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS
384 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS
386 If you say Y here the resulting kernel image will be slightly
387 larger and slower, but it gives very useful debugging information
388 in case of kernel bugs. (precise oopses/stacktraces/warnings)
390 config STACK_VALIDATION
391 bool "Compile-time stack metadata validation"
392 depends on HAVE_STACK_VALIDATION
395 Add compile-time checks to validate stack metadata, including frame
396 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure
397 that runtime stack traces are more reliable.
399 This is also a prerequisite for generation of ORC unwind data, which
400 is needed for CONFIG_UNWINDER_ORC.
402 For more information, see
403 tools/objtool/Documentation/stack-validation.txt.
405 config DEBUG_FORCE_WEAK_PER_CPU
406 bool "Force weak per-cpu definitions"
407 depends on DEBUG_KERNEL
409 s390 and alpha require percpu variables in modules to be
410 defined weak to work around addressing range issue which
411 puts the following two restrictions on percpu variable
414 1. percpu symbols must be unique whether static or not
415 2. percpu variables can't be defined inside a function
417 To ensure that generic code follows the above rules, this
418 option forces all percpu variables to be defined as weak.
420 endmenu # "Compiler options"
423 bool "Magic SysRq key"
426 If you say Y here, you will have some control over the system even
427 if the system crashes for example during kernel debugging (e.g., you
428 will be able to flush the buffer cache to disk, reboot the system
429 immediately or dump some status information). This is accomplished
430 by pressing various keys while holding SysRq (Alt+PrintScreen). It
431 also works on a serial console (on PC hardware at least), if you
432 send a BREAK and then within 5 seconds a command keypress. The
433 keys are documented in <file:Documentation/admin-guide/sysrq.rst>.
434 Don't say Y unless you really know what this hack does.
436 config MAGIC_SYSRQ_DEFAULT_ENABLE
437 hex "Enable magic SysRq key functions by default"
438 depends on MAGIC_SYSRQ
441 Specifies which SysRq key functions are enabled by default.
442 This may be set to 1 or 0 to enable or disable them all, or
443 to a bitmask as described in Documentation/admin-guide/sysrq.rst.
445 config MAGIC_SYSRQ_SERIAL
446 bool "Enable magic SysRq key over serial"
447 depends on MAGIC_SYSRQ
450 Many embedded boards have a disconnected TTL level serial which can
451 generate some garbage that can lead to spurious false sysrq detects.
452 This option allows you to decide whether you want to enable the
456 bool "Kernel debugging"
458 Say Y here if you are developing drivers or trying to debug and
459 identify kernel problems.
462 bool "Miscellaneous debug code"
464 depends on DEBUG_KERNEL
466 Say Y here if you need to enable miscellaneous debug code that should
467 be under a more specific debug option but isn't.
470 menu "Memory Debugging"
472 source "mm/Kconfig.debug"
475 bool "Debug object operations"
476 depends on DEBUG_KERNEL
478 If you say Y here, additional code will be inserted into the
479 kernel to track the life time of various objects and validate
480 the operations on those objects.
482 config DEBUG_OBJECTS_SELFTEST
483 bool "Debug objects selftest"
484 depends on DEBUG_OBJECTS
486 This enables the selftest of the object debug code.
488 config DEBUG_OBJECTS_FREE
489 bool "Debug objects in freed memory"
490 depends on DEBUG_OBJECTS
492 This enables checks whether a k/v free operation frees an area
493 which contains an object which has not been deactivated
494 properly. This can make kmalloc/kfree-intensive workloads
497 config DEBUG_OBJECTS_TIMERS
498 bool "Debug timer objects"
499 depends on DEBUG_OBJECTS
501 If you say Y here, additional code will be inserted into the
502 timer routines to track the life time of timer objects and
503 validate the timer operations.
505 config DEBUG_OBJECTS_WORK
506 bool "Debug work objects"
507 depends on DEBUG_OBJECTS
509 If you say Y here, additional code will be inserted into the
510 work queue routines to track the life time of work objects and
511 validate the work operations.
513 config DEBUG_OBJECTS_RCU_HEAD
514 bool "Debug RCU callbacks objects"
515 depends on DEBUG_OBJECTS
517 Enable this to turn on debugging of RCU list heads (call_rcu() usage).
519 config DEBUG_OBJECTS_PERCPU_COUNTER
520 bool "Debug percpu counter objects"
521 depends on DEBUG_OBJECTS
523 If you say Y here, additional code will be inserted into the
524 percpu counter routines to track the life time of percpu counter
525 objects and validate the percpu counter operations.
527 config DEBUG_OBJECTS_ENABLE_DEFAULT
528 int "debug_objects bootup default value (0-1)"
531 depends on DEBUG_OBJECTS
533 Debug objects boot parameter default value
536 bool "Debug slab memory allocations"
537 depends on DEBUG_KERNEL && SLAB
539 Say Y here to have the kernel do limited verification on memory
540 allocation as well as poisoning memory on free to catch use of freed
541 memory. This can make kmalloc/kfree-intensive workloads much slower.
544 bool "SLUB debugging on by default"
545 depends on SLUB && SLUB_DEBUG
548 Boot with debugging on by default. SLUB boots by default with
549 the runtime debug capabilities switched off. Enabling this is
550 equivalent to specifying the "slub_debug" parameter on boot.
551 There is no support for more fine grained debug control like
552 possible with slub_debug=xxx. SLUB debugging may be switched
553 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying
558 bool "Enable SLUB performance statistics"
559 depends on SLUB && SYSFS
561 SLUB statistics are useful to debug SLUBs allocation behavior in
562 order find ways to optimize the allocator. This should never be
563 enabled for production use since keeping statistics slows down
564 the allocator by a few percentage points. The slabinfo command
565 supports the determination of the most active slabs to figure
566 out which slabs are relevant to a particular load.
567 Try running: slabinfo -DA
569 config HAVE_DEBUG_KMEMLEAK
572 config DEBUG_KMEMLEAK
573 bool "Kernel memory leak detector"
574 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK
576 select STACKTRACE if STACKTRACE_SUPPORT
580 Say Y here if you want to enable the memory leak
581 detector. The memory allocation/freeing is traced in a way
582 similar to the Boehm's conservative garbage collector, the
583 difference being that the orphan objects are not freed but
584 only shown in /sys/kernel/debug/kmemleak. Enabling this
585 feature will introduce an overhead to memory
586 allocations. See Documentation/dev-tools/kmemleak.rst for more
589 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
590 of finding leaks due to the slab objects poisoning.
592 In order to access the kmemleak file, debugfs needs to be
593 mounted (usually at /sys/kernel/debug).
595 config DEBUG_KMEMLEAK_EARLY_LOG_SIZE
596 int "Maximum kmemleak early log entries"
597 depends on DEBUG_KMEMLEAK
601 Kmemleak must track all the memory allocations to avoid
602 reporting false positives. Since memory may be allocated or
603 freed before kmemleak is initialised, an early log buffer is
604 used to store these actions. If kmemleak reports "early log
605 buffer exceeded", please increase this value.
607 config DEBUG_KMEMLEAK_TEST
608 tristate "Simple test for the kernel memory leak detector"
609 depends on DEBUG_KMEMLEAK && m
611 This option enables a module that explicitly leaks memory.
615 config DEBUG_KMEMLEAK_DEFAULT_OFF
616 bool "Default kmemleak to off"
617 depends on DEBUG_KMEMLEAK
619 Say Y here to disable kmemleak by default. It can then be enabled
620 on the command line via kmemleak=on.
622 config DEBUG_KMEMLEAK_AUTO_SCAN
623 bool "Enable kmemleak auto scan thread on boot up"
625 depends on DEBUG_KMEMLEAK
627 Depending on the cpu, kmemleak scan may be cpu intensive and can
628 stall user tasks at times. This option enables/disables automatic
629 kmemleak scan at boot up.
631 Say N here to disable kmemleak auto scan thread to stop automatic
632 scanning. Disabling this option disables automatic reporting of
637 config DEBUG_STACK_USAGE
638 bool "Stack utilization instrumentation"
639 depends on DEBUG_KERNEL && !IA64
641 Enables the display of the minimum amount of free stack which each
642 task has ever had available in the sysrq-T and sysrq-P debug output.
644 This option will slow down process creation somewhat.
648 depends on DEBUG_KERNEL
650 Enable this to turn on extended checks in the virtual-memory system
651 that may impact performance.
655 config DEBUG_VM_VMACACHE
656 bool "Debug VMA caching"
659 Enable this to turn on VMA caching debug information. Doing so
660 can cause significant overhead, so only enable it in non-production
666 bool "Debug VM red-black trees"
669 Enable VM red-black tree debugging information and extra validations.
673 config DEBUG_VM_PGFLAGS
674 bool "Debug page-flags operations"
677 Enables extra validation on page flags operations.
681 config ARCH_HAS_DEBUG_VIRTUAL
685 bool "Debug VM translations"
686 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL
688 Enable some costly sanity checks in virtual to page code. This can
689 catch mistakes with virt_to_page() and friends.
693 config DEBUG_NOMMU_REGIONS
694 bool "Debug the global anon/private NOMMU mapping region tree"
695 depends on DEBUG_KERNEL && !MMU
697 This option causes the global tree of anonymous and private mapping
698 regions to be regularly checked for invalid topology.
700 config DEBUG_MEMORY_INIT
701 bool "Debug memory initialisation" if EXPERT
704 Enable this for additional checks during memory initialisation.
705 The sanity checks verify aspects of the VM such as the memory model
706 and other information provided by the architecture. Verbose
707 information will be printed at KERN_DEBUG loglevel depending
708 on the mminit_loglevel= command-line option.
712 config MEMORY_NOTIFIER_ERROR_INJECT
713 tristate "Memory hotplug notifier error injection module"
714 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION
716 This option provides the ability to inject artificial errors to
717 memory hotplug notifier chain callbacks. It is controlled through
718 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory
720 If the notifier call chain should be failed with some events
721 notified, write the error code to "actions/<notifier event>/error".
723 Example: Inject memory hotplug offline error (-12 == -ENOMEM)
725 # cd /sys/kernel/debug/notifier-error-inject/memory
726 # echo -12 > actions/MEM_GOING_OFFLINE/error
727 # echo offline > /sys/devices/system/memory/memoryXXX/state
728 bash: echo: write error: Cannot allocate memory
730 To compile this code as a module, choose M here: the module will
731 be called memory-notifier-error-inject.
735 config DEBUG_PER_CPU_MAPS
736 bool "Debug access to per_cpu maps"
737 depends on DEBUG_KERNEL
740 Say Y to verify that the per_cpu map being accessed has
741 been set up. This adds a fair amount of code to kernel memory
742 and decreases performance.
747 bool "Highmem debugging"
748 depends on DEBUG_KERNEL && HIGHMEM
750 This option enables additional error checking for high memory
751 systems. Disable for production systems.
753 config HAVE_DEBUG_STACKOVERFLOW
756 config DEBUG_STACKOVERFLOW
757 bool "Check for stack overflows"
758 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW
760 Say Y here if you want to check for overflows of kernel, IRQ
761 and exception stacks (if your architecture uses them). This
762 option will show detailed messages if free stack space drops
763 below a certain limit.
765 These kinds of bugs usually occur when call-chains in the
766 kernel get too deep, especially when interrupts are
769 Use this in cases where you see apparently random memory
770 corruption, especially if it appears in 'struct thread_info'
772 If in doubt, say "N".
774 source "lib/Kconfig.kasan"
776 endmenu # "Memory Debugging"
781 An architecture should select this when it can successfully
782 build and run with CONFIG_KCOV. This typically requires
783 disabling instrumentation for some early boot code.
785 config CC_HAS_SANCOV_TRACE_PC
786 def_bool $(cc-option,-fsanitize-coverage=trace-pc)
789 bool "Code coverage for fuzzing"
790 depends on ARCH_HAS_KCOV
791 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS
793 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC
795 KCOV exposes kernel code coverage information in a form suitable
796 for coverage-guided fuzzing (randomized testing).
798 If RANDOMIZE_BASE is enabled, PC values will not be stable across
799 different machines and across reboots. If you need stable PC values,
800 disable RANDOMIZE_BASE.
802 For more details, see Documentation/dev-tools/kcov.rst.
804 config KCOV_ENABLE_COMPARISONS
805 bool "Enable comparison operands collection by KCOV"
807 depends on $(cc-option,-fsanitize-coverage=trace-cmp)
809 KCOV also exposes operands of every comparison in the instrumented
810 code along with operand sizes and PCs of the comparison instructions.
811 These operands can be used by fuzzing engines to improve the quality
814 config KCOV_INSTRUMENT_ALL
815 bool "Instrument all code by default"
819 If you are doing generic system call fuzzing (like e.g. syzkaller),
820 then you will want to instrument the whole kernel and you should
821 say y here. If you are doing more targeted fuzzing (like e.g.
822 filesystem fuzzing with AFL) then you will want to enable coverage
823 for more specific subsets of files, and should say n here.
826 bool "Debug shared IRQ handlers"
827 depends on DEBUG_KERNEL
829 Enable this to generate a spurious interrupt as soon as a shared
830 interrupt handler is registered, and just before one is deregistered.
831 Drivers ought to be able to handle interrupts coming in at those
832 points; some don't and need to be caught.
834 menu "Debug Lockups and Hangs"
836 config LOCKUP_DETECTOR
839 config SOFTLOCKUP_DETECTOR
840 bool "Detect Soft Lockups"
841 depends on DEBUG_KERNEL && !S390
842 select LOCKUP_DETECTOR
844 Say Y here to enable the kernel to act as a watchdog to detect
847 Softlockups are bugs that cause the kernel to loop in kernel
848 mode for more than 20 seconds, without giving other tasks a
849 chance to run. The current stack trace is displayed upon
850 detection and the system will stay locked up.
852 config BOOTPARAM_SOFTLOCKUP_PANIC
853 bool "Panic (Reboot) On Soft Lockups"
854 depends on SOFTLOCKUP_DETECTOR
856 Say Y here to enable the kernel to panic on "soft lockups",
857 which are bugs that cause the kernel to loop in kernel
858 mode for more than 20 seconds (configurable using the watchdog_thresh
859 sysctl), without giving other tasks a chance to run.
861 The panic can be used in combination with panic_timeout,
862 to cause the system to reboot automatically after a
863 lockup has been detected. This feature is useful for
864 high-availability systems that have uptime guarantees and
865 where a lockup must be resolved ASAP.
869 config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE
871 depends on SOFTLOCKUP_DETECTOR
873 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC
874 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC
876 config HARDLOCKUP_DETECTOR_PERF
878 select SOFTLOCKUP_DETECTOR
881 # Enables a timestamp based low pass filter to compensate for perf based
882 # hard lockup detection which runs too fast due to turbo modes.
884 config HARDLOCKUP_CHECK_TIMESTAMP
888 # arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard
889 # lockup detector rather than the perf based detector.
891 config HARDLOCKUP_DETECTOR
892 bool "Detect Hard Lockups"
893 depends on DEBUG_KERNEL && !S390
894 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH
895 select LOCKUP_DETECTOR
896 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF
897 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH
899 Say Y here to enable the kernel to act as a watchdog to detect
902 Hardlockups are bugs that cause the CPU to loop in kernel mode
903 for more than 10 seconds, without letting other interrupts have a
904 chance to run. The current stack trace is displayed upon detection
905 and the system will stay locked up.
907 config BOOTPARAM_HARDLOCKUP_PANIC
908 bool "Panic (Reboot) On Hard Lockups"
909 depends on HARDLOCKUP_DETECTOR
911 Say Y here to enable the kernel to panic on "hard lockups",
912 which are bugs that cause the kernel to loop in kernel
913 mode with interrupts disabled for more than 10 seconds (configurable
914 using the watchdog_thresh sysctl).
918 config BOOTPARAM_HARDLOCKUP_PANIC_VALUE
920 depends on HARDLOCKUP_DETECTOR
922 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC
923 default 1 if BOOTPARAM_HARDLOCKUP_PANIC
925 config DETECT_HUNG_TASK
926 bool "Detect Hung Tasks"
927 depends on DEBUG_KERNEL
928 default SOFTLOCKUP_DETECTOR
930 Say Y here to enable the kernel to detect "hung tasks",
931 which are bugs that cause the task to be stuck in
932 uninterruptible "D" state indefinitely.
934 When a hung task is detected, the kernel will print the
935 current stack trace (which you should report), but the
936 task will stay in uninterruptible state. If lockdep is
937 enabled then all held locks will also be reported. This
938 feature has negligible overhead.
940 config DEFAULT_HUNG_TASK_TIMEOUT
941 int "Default timeout for hung task detection (in seconds)"
942 depends on DETECT_HUNG_TASK
945 This option controls the default timeout (in seconds) used
946 to determine when a task has become non-responsive and should
949 It can be adjusted at runtime via the kernel.hung_task_timeout_secs
950 sysctl or by writing a value to
951 /proc/sys/kernel/hung_task_timeout_secs.
953 A timeout of 0 disables the check. The default is two minutes.
954 Keeping the default should be fine in most cases.
956 config BOOTPARAM_HUNG_TASK_PANIC
957 bool "Panic (Reboot) On Hung Tasks"
958 depends on DETECT_HUNG_TASK
960 Say Y here to enable the kernel to panic on "hung tasks",
961 which are bugs that cause the kernel to leave a task stuck
962 in uninterruptible "D" state.
964 The panic can be used in combination with panic_timeout,
965 to cause the system to reboot automatically after a
966 hung task has been detected. This feature is useful for
967 high-availability systems that have uptime guarantees and
968 where a hung tasks must be resolved ASAP.
972 config BOOTPARAM_HUNG_TASK_PANIC_VALUE
974 depends on DETECT_HUNG_TASK
976 default 0 if !BOOTPARAM_HUNG_TASK_PANIC
977 default 1 if BOOTPARAM_HUNG_TASK_PANIC
980 bool "Detect Workqueue Stalls"
981 depends on DEBUG_KERNEL
983 Say Y here to enable stall detection on workqueues. If a
984 worker pool doesn't make forward progress on a pending work
985 item for over a given amount of time, 30s by default, a
986 warning message is printed along with dump of workqueue
987 state. This can be configured through kernel parameter
988 "workqueue.watchdog_thresh" and its sysfs counterpart.
990 endmenu # "Debug lockups and hangs"
995 Say Y here to enable the kernel to panic when it oopses. This
996 has the same effect as setting oops=panic on the kernel command
999 This feature is useful to ensure that the kernel does not do
1000 anything erroneous after an oops which could result in data
1001 corruption or other issues.
1005 config PANIC_ON_OOPS_VALUE
1008 default 0 if !PANIC_ON_OOPS
1009 default 1 if PANIC_ON_OOPS
1011 config PANIC_TIMEOUT
1015 Set the timeout value (in seconds) until a reboot occurs when the
1016 the kernel panics. If n = 0, then we wait forever. A timeout
1017 value n > 0 will wait n seconds before rebooting, while a timeout
1018 value n < 0 will reboot immediately.
1021 bool "Collect scheduler debugging info"
1022 depends on DEBUG_KERNEL && PROC_FS
1025 If you say Y here, the /proc/sched_debug file will be provided
1026 that can help debug the scheduler. The runtime overhead of this
1034 bool "Collect scheduler statistics"
1035 depends on DEBUG_KERNEL && PROC_FS
1038 If you say Y here, additional code will be inserted into the
1039 scheduler and related routines to collect statistics about
1040 scheduler behavior and provide them in /proc/schedstat. These
1041 stats may be useful for both tuning and debugging the scheduler
1042 If you aren't debugging the scheduler or trying to tune a specific
1043 application, you can say N to avoid the very slight overhead
1046 config SCHED_STACK_END_CHECK
1047 bool "Detect stack corruption on calls to schedule()"
1048 depends on DEBUG_KERNEL
1051 This option checks for a stack overrun on calls to schedule().
1052 If the stack end location is found to be over written always panic as
1053 the content of the corrupted region can no longer be trusted.
1054 This is to ensure no erroneous behaviour occurs which could result in
1055 data corruption or a sporadic crash at a later stage once the region
1056 is examined. The runtime overhead introduced is minimal.
1058 config DEBUG_TIMEKEEPING
1059 bool "Enable extra timekeeping sanity checking"
1061 This option will enable additional timekeeping sanity checks
1062 which may be helpful when diagnosing issues where timekeeping
1063 problems are suspected.
1065 This may include checks in the timekeeping hotpaths, so this
1066 option may have a (very small) performance impact to some
1071 config DEBUG_PREEMPT
1072 bool "Debug preemptible kernel"
1073 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT
1076 If you say Y here then the kernel will use a debug variant of the
1077 commonly used smp_processor_id() function and will print warnings
1078 if kernel code uses it in a preemption-unsafe way. Also, the kernel
1079 will detect preemption count underflows.
1081 menu "Lock Debugging (spinlocks, mutexes, etc...)"
1083 config LOCK_DEBUGGING_SUPPORT
1085 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
1088 config PROVE_LOCKING
1089 bool "Lock debugging: prove locking correctness"
1090 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1092 select DEBUG_SPINLOCK
1093 select DEBUG_MUTEXES
1094 select DEBUG_RT_MUTEXES if RT_MUTEXES
1096 select DEBUG_WW_MUTEX_SLOWPATH
1097 select DEBUG_LOCK_ALLOC
1098 select TRACE_IRQFLAGS
1101 This feature enables the kernel to prove that all locking
1102 that occurs in the kernel runtime is mathematically
1103 correct: that under no circumstance could an arbitrary (and
1104 not yet triggered) combination of observed locking
1105 sequences (on an arbitrary number of CPUs, running an
1106 arbitrary number of tasks and interrupt contexts) cause a
1109 In short, this feature enables the kernel to report locking
1110 related deadlocks before they actually occur.
1112 The proof does not depend on how hard and complex a
1113 deadlock scenario would be to trigger: how many
1114 participant CPUs, tasks and irq-contexts would be needed
1115 for it to trigger. The proof also does not depend on
1116 timing: if a race and a resulting deadlock is possible
1117 theoretically (no matter how unlikely the race scenario
1118 is), it will be proven so and will immediately be
1119 reported by the kernel (once the event is observed that
1120 makes the deadlock theoretically possible).
1122 If a deadlock is impossible (i.e. the locking rules, as
1123 observed by the kernel, are mathematically correct), the
1124 kernel reports nothing.
1126 NOTE: this feature can also be enabled for rwlocks, mutexes
1127 and rwsems - in which case all dependencies between these
1128 different locking variants are observed and mapped too, and
1129 the proof of observed correctness is also maintained for an
1130 arbitrary combination of these separate locking variants.
1132 For more details, see Documentation/locking/lockdep-design.rst.
1135 bool "Lock usage statistics"
1136 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1138 select DEBUG_SPINLOCK
1139 select DEBUG_MUTEXES
1140 select DEBUG_RT_MUTEXES if RT_MUTEXES
1141 select DEBUG_LOCK_ALLOC
1144 This feature enables tracking lock contention points
1146 For more details, see Documentation/locking/lockstat.rst
1148 This also enables lock events required by "perf lock",
1150 If you want to use "perf lock", you also need to turn on
1151 CONFIG_EVENT_TRACING.
1153 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events.
1154 (CONFIG_LOCKDEP defines "acquire" and "release" events.)
1156 config DEBUG_RT_MUTEXES
1157 bool "RT Mutex debugging, deadlock detection"
1158 depends on DEBUG_KERNEL && RT_MUTEXES
1160 This allows rt mutex semantics violations and rt mutex related
1161 deadlocks (lockups) to be detected and reported automatically.
1163 config DEBUG_SPINLOCK
1164 bool "Spinlock and rw-lock debugging: basic checks"
1165 depends on DEBUG_KERNEL
1166 select UNINLINE_SPIN_UNLOCK
1168 Say Y here and build SMP to catch missing spinlock initialization
1169 and certain other kinds of spinlock errors commonly made. This is
1170 best used in conjunction with the NMI watchdog so that spinlock
1171 deadlocks are also debuggable.
1173 config DEBUG_MUTEXES
1174 bool "Mutex debugging: basic checks"
1175 depends on DEBUG_KERNEL
1177 This feature allows mutex semantics violations to be detected and
1180 config DEBUG_WW_MUTEX_SLOWPATH
1181 bool "Wait/wound mutex debugging: Slowpath testing"
1182 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1183 select DEBUG_LOCK_ALLOC
1184 select DEBUG_SPINLOCK
1185 select DEBUG_MUTEXES
1187 This feature enables slowpath testing for w/w mutex users by
1188 injecting additional -EDEADLK wound/backoff cases. Together with
1189 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this
1190 will test all possible w/w mutex interface abuse with the
1191 exception of simply not acquiring all the required locks.
1192 Note that this feature can introduce significant overhead, so
1193 it really should not be enabled in a production or distro kernel,
1194 even a debug kernel. If you are a driver writer, enable it. If
1195 you are a distro, do not.
1198 bool "RW Semaphore debugging: basic checks"
1199 depends on DEBUG_KERNEL
1201 This debugging feature allows mismatched rw semaphore locks
1202 and unlocks to be detected and reported.
1204 config DEBUG_LOCK_ALLOC
1205 bool "Lock debugging: detect incorrect freeing of live locks"
1206 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1207 select DEBUG_SPINLOCK
1208 select DEBUG_MUTEXES
1209 select DEBUG_RT_MUTEXES if RT_MUTEXES
1212 This feature will check whether any held lock (spinlock, rwlock,
1213 mutex or rwsem) is incorrectly freed by the kernel, via any of the
1214 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
1215 vfree(), etc.), whether a live lock is incorrectly reinitialized via
1216 spin_lock_init()/mutex_init()/etc., or whether there is any lock
1217 held during task exit.
1221 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
1223 select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86
1227 config LOCKDEP_SMALL
1230 config DEBUG_LOCKDEP
1231 bool "Lock dependency engine debugging"
1232 depends on DEBUG_KERNEL && LOCKDEP
1234 If you say Y here, the lock dependency engine will do
1235 additional runtime checks to debug itself, at the price
1236 of more runtime overhead.
1238 config DEBUG_ATOMIC_SLEEP
1239 bool "Sleep inside atomic section checking"
1240 select PREEMPT_COUNT
1241 depends on DEBUG_KERNEL
1242 depends on !ARCH_NO_PREEMPT
1244 If you say Y here, various routines which may sleep will become very
1245 noisy if they are called inside atomic sections: when a spinlock is
1246 held, inside an rcu read side critical section, inside preempt disabled
1247 sections, inside an interrupt, etc...
1249 config DEBUG_LOCKING_API_SELFTESTS
1250 bool "Locking API boot-time self-tests"
1251 depends on DEBUG_KERNEL
1253 Say Y here if you want the kernel to run a short self-test during
1254 bootup. The self-test checks whether common types of locking bugs
1255 are detected by debugging mechanisms or not. (if you disable
1256 lock debugging then those bugs wont be detected of course.)
1257 The following locking APIs are covered: spinlocks, rwlocks,
1260 config LOCK_TORTURE_TEST
1261 tristate "torture tests for locking"
1262 depends on DEBUG_KERNEL
1265 This option provides a kernel module that runs torture tests
1266 on kernel locking primitives. The kernel module may be built
1267 after the fact on the running kernel to be tested, if desired.
1269 Say Y here if you want kernel locking-primitive torture tests
1270 to be built into the kernel.
1271 Say M if you want these torture tests to build as a module.
1272 Say N if you are unsure.
1274 config WW_MUTEX_SELFTEST
1275 tristate "Wait/wound mutex selftests"
1277 This option provides a kernel module that runs tests on the
1278 on the struct ww_mutex locking API.
1280 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction
1281 with this test harness.
1283 Say M if you want these self tests to build as a module.
1284 Say N if you are unsure.
1286 endmenu # lock debugging
1288 config TRACE_IRQFLAGS
1291 Enables hooks to interrupt enabling and disabling for
1292 either tracing or lock debugging.
1295 bool "Stack backtrace support"
1296 depends on STACKTRACE_SUPPORT
1298 This option causes the kernel to create a /proc/pid/stack for
1299 every process, showing its current stack trace.
1300 It is also used by various kernel debugging features that require
1301 stack trace generation.
1303 config WARN_ALL_UNSEEDED_RANDOM
1304 bool "Warn for all uses of unseeded randomness"
1307 Some parts of the kernel contain bugs relating to their use of
1308 cryptographically secure random numbers before it's actually possible
1309 to generate those numbers securely. This setting ensures that these
1310 flaws don't go unnoticed, by enabling a message, should this ever
1311 occur. This will allow people with obscure setups to know when things
1312 are going wrong, so that they might contact developers about fixing
1315 Unfortunately, on some models of some architectures getting
1316 a fully seeded CRNG is extremely difficult, and so this can
1317 result in dmesg getting spammed for a surprisingly long
1318 time. This is really bad from a security perspective, and
1319 so architecture maintainers really need to do what they can
1320 to get the CRNG seeded sooner after the system is booted.
1321 However, since users cannot do anything actionable to
1322 address this, by default the kernel will issue only a single
1323 warning for the first use of unseeded randomness.
1325 Say Y here if you want to receive warnings for all uses of
1326 unseeded randomness. This will be of use primarily for
1327 those developers interested in improving the security of
1328 Linux kernels running on their architecture (or
1331 config DEBUG_KOBJECT
1332 bool "kobject debugging"
1333 depends on DEBUG_KERNEL
1335 If you say Y here, some extra kobject debugging messages will be sent
1338 config DEBUG_KOBJECT_RELEASE
1339 bool "kobject release debugging"
1340 depends on DEBUG_OBJECTS_TIMERS
1342 kobjects are reference counted objects. This means that their
1343 last reference count put is not predictable, and the kobject can
1344 live on past the point at which a driver decides to drop it's
1345 initial reference to the kobject gained on allocation. An
1346 example of this would be a struct device which has just been
1349 However, some buggy drivers assume that after such an operation,
1350 the memory backing the kobject can be immediately freed. This
1351 goes completely against the principles of a refcounted object.
1353 If you say Y here, the kernel will delay the release of kobjects
1354 on the last reference count to improve the visibility of this
1355 kind of kobject release bug.
1357 config HAVE_DEBUG_BUGVERBOSE
1360 config DEBUG_BUGVERBOSE
1361 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT
1362 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE)
1365 Say Y here to make BUG() panics output the file name and line number
1366 of the BUG call as well as the EIP and oops trace. This aids
1367 debugging but costs about 70-100K of memory.
1370 bool "Debug linked list manipulation"
1371 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION
1373 Enable this to turn on extended checks in the linked-list
1379 bool "Debug priority linked list manipulation"
1380 depends on DEBUG_KERNEL
1382 Enable this to turn on extended checks in the priority-ordered
1383 linked-list (plist) walking routines. This checks the entire
1384 list multiple times during each manipulation.
1389 bool "Debug SG table operations"
1390 depends on DEBUG_KERNEL
1392 Enable this to turn on checks on scatter-gather tables. This can
1393 help find problems with drivers that do not properly initialize
1398 config DEBUG_NOTIFIERS
1399 bool "Debug notifier call chains"
1400 depends on DEBUG_KERNEL
1402 Enable this to turn on sanity checking for notifier call chains.
1403 This is most useful for kernel developers to make sure that
1404 modules properly unregister themselves from notifier chains.
1405 This is a relatively cheap check but if you care about maximum
1408 config DEBUG_CREDENTIALS
1409 bool "Debug credential management"
1410 depends on DEBUG_KERNEL
1412 Enable this to turn on some debug checking for credential
1413 management. The additional code keeps track of the number of
1414 pointers from task_structs to any given cred struct, and checks to
1415 see that this number never exceeds the usage count of the cred
1418 Furthermore, if SELinux is enabled, this also checks that the
1419 security pointer in the cred struct is never seen to be invalid.
1423 source "kernel/rcu/Kconfig.debug"
1425 config DEBUG_WQ_FORCE_RR_CPU
1426 bool "Force round-robin CPU selection for unbound work items"
1427 depends on DEBUG_KERNEL
1430 Workqueue used to implicitly guarantee that work items queued
1431 without explicit CPU specified are put on the local CPU. This
1432 guarantee is no longer true and while local CPU is still
1433 preferred work items may be put on foreign CPUs. Kernel
1434 parameter "workqueue.debug_force_rr_cpu" is added to force
1435 round-robin CPU selection to flush out usages which depend on the
1436 now broken guarantee. This config option enables the debug
1437 feature by default. When enabled, memory and cache locality will
1440 config DEBUG_BLOCK_EXT_DEVT
1441 bool "Force extended block device numbers and spread them"
1442 depends on DEBUG_KERNEL
1446 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON
1447 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT
1448 YOU ARE DOING. Distros, please enable this and fix whatever
1451 Conventionally, block device numbers are allocated from
1452 predetermined contiguous area. However, extended block area
1453 may introduce non-contiguous block device numbers. This
1454 option forces most block device numbers to be allocated from
1455 the extended space and spreads them to discover kernel or
1456 userland code paths which assume predetermined contiguous
1457 device number allocation.
1459 Note that turning on this debug option shuffles all the
1460 device numbers for all IDE and SCSI devices including libata
1461 ones, so root partition specified using device number
1462 directly (via rdev or root=MAJ:MIN) won't work anymore.
1463 Textual device names (root=/dev/sdXn) will continue to work.
1465 Say N if you are unsure.
1467 config CPU_HOTPLUG_STATE_CONTROL
1468 bool "Enable CPU hotplug state control"
1469 depends on DEBUG_KERNEL
1470 depends on HOTPLUG_CPU
1473 Allows to write steps between "offline" and "online" to the CPUs
1474 sysfs target file so states can be stepped granular. This is a debug
1475 option for now as the hotplug machinery cannot be stopped and
1476 restarted at arbitrary points yet.
1478 Say N if your are unsure.
1480 config NOTIFIER_ERROR_INJECTION
1481 tristate "Notifier error injection"
1482 depends on DEBUG_KERNEL
1485 This option provides the ability to inject artificial errors to
1486 specified notifier chain callbacks. It is useful to test the error
1487 handling of notifier call chain failures.
1491 config PM_NOTIFIER_ERROR_INJECT
1492 tristate "PM notifier error injection module"
1493 depends on PM && NOTIFIER_ERROR_INJECTION
1494 default m if PM_DEBUG
1496 This option provides the ability to inject artificial errors to
1497 PM notifier chain callbacks. It is controlled through debugfs
1498 interface /sys/kernel/debug/notifier-error-inject/pm
1500 If the notifier call chain should be failed with some events
1501 notified, write the error code to "actions/<notifier event>/error".
1503 Example: Inject PM suspend error (-12 = -ENOMEM)
1505 # cd /sys/kernel/debug/notifier-error-inject/pm/
1506 # echo -12 > actions/PM_SUSPEND_PREPARE/error
1507 # echo mem > /sys/power/state
1508 bash: echo: write error: Cannot allocate memory
1510 To compile this code as a module, choose M here: the module will
1511 be called pm-notifier-error-inject.
1515 config OF_RECONFIG_NOTIFIER_ERROR_INJECT
1516 tristate "OF reconfig notifier error injection module"
1517 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION
1519 This option provides the ability to inject artificial errors to
1520 OF reconfig notifier chain callbacks. It is controlled
1521 through debugfs interface under
1522 /sys/kernel/debug/notifier-error-inject/OF-reconfig/
1524 If the notifier call chain should be failed with some events
1525 notified, write the error code to "actions/<notifier event>/error".
1527 To compile this code as a module, choose M here: the module will
1528 be called of-reconfig-notifier-error-inject.
1532 config NETDEV_NOTIFIER_ERROR_INJECT
1533 tristate "Netdev notifier error injection module"
1534 depends on NET && NOTIFIER_ERROR_INJECTION
1536 This option provides the ability to inject artificial errors to
1537 netdevice notifier chain callbacks. It is controlled through debugfs
1538 interface /sys/kernel/debug/notifier-error-inject/netdev
1540 If the notifier call chain should be failed with some events
1541 notified, write the error code to "actions/<notifier event>/error".
1543 Example: Inject netdevice mtu change error (-22 = -EINVAL)
1545 # cd /sys/kernel/debug/notifier-error-inject/netdev
1546 # echo -22 > actions/NETDEV_CHANGEMTU/error
1547 # ip link set eth0 mtu 1024
1548 RTNETLINK answers: Invalid argument
1550 To compile this code as a module, choose M here: the module will
1551 be called netdev-notifier-error-inject.
1555 config FUNCTION_ERROR_INJECTION
1557 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES
1559 config FAULT_INJECTION
1560 bool "Fault-injection framework"
1561 depends on DEBUG_KERNEL
1563 Provide fault-injection framework.
1564 For more details, see Documentation/fault-injection/.
1567 bool "Fault-injection capability for kmalloc"
1568 depends on FAULT_INJECTION
1569 depends on SLAB || SLUB
1571 Provide fault-injection capability for kmalloc.
1573 config FAIL_PAGE_ALLOC
1574 bool "Fault-injection capabilitiy for alloc_pages()"
1575 depends on FAULT_INJECTION
1577 Provide fault-injection capability for alloc_pages().
1579 config FAIL_MAKE_REQUEST
1580 bool "Fault-injection capability for disk IO"
1581 depends on FAULT_INJECTION && BLOCK
1583 Provide fault-injection capability for disk IO.
1585 config FAIL_IO_TIMEOUT
1586 bool "Fault-injection capability for faking disk interrupts"
1587 depends on FAULT_INJECTION && BLOCK
1589 Provide fault-injection capability on end IO handling. This
1590 will make the block layer "forget" an interrupt as configured,
1591 thus exercising the error handling.
1593 Only works with drivers that use the generic timeout handling,
1594 for others it wont do anything.
1597 bool "Fault-injection capability for futexes"
1599 depends on FAULT_INJECTION && FUTEX
1601 Provide fault-injection capability for futexes.
1603 config FAULT_INJECTION_DEBUG_FS
1604 bool "Debugfs entries for fault-injection capabilities"
1605 depends on FAULT_INJECTION && SYSFS && DEBUG_FS
1607 Enable configuration of fault-injection capabilities via debugfs.
1609 config FAIL_FUNCTION
1610 bool "Fault-injection capability for functions"
1611 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION
1613 Provide function-based fault-injection capability.
1614 This will allow you to override a specific function with a return
1615 with given return value. As a result, function caller will see
1616 an error value and have to handle it. This is useful to test the
1617 error handling in various subsystems.
1619 config FAIL_MMC_REQUEST
1620 bool "Fault-injection capability for MMC IO"
1621 depends on FAULT_INJECTION_DEBUG_FS && MMC
1623 Provide fault-injection capability for MMC IO.
1624 This will make the mmc core return data errors. This is
1625 useful to test the error handling in the mmc block device
1626 and to test how the mmc host driver handles retries from
1629 config FAULT_INJECTION_STACKTRACE_FILTER
1630 bool "stacktrace filter for fault-injection capabilities"
1631 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT
1634 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86
1636 Provide stacktrace filter for fault-injection capabilities
1639 bool "Latency measuring infrastructure"
1640 depends on DEBUG_KERNEL
1641 depends on STACKTRACE_SUPPORT
1643 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86
1650 Enable this option if you want to use the LatencyTOP tool
1651 to find out which userspace is blocking on what kernel operations.
1653 source "kernel/trace/Kconfig"
1655 config PROVIDE_OHCI1394_DMA_INIT
1656 bool "Remote debugging over FireWire early on boot"
1657 depends on PCI && X86
1659 If you want to debug problems which hang or crash the kernel early
1660 on boot and the crashing machine has a FireWire port, you can use
1661 this feature to remotely access the memory of the crashed machine
1662 over FireWire. This employs remote DMA as part of the OHCI1394
1663 specification which is now the standard for FireWire controllers.
1665 With remote DMA, you can monitor the printk buffer remotely using
1666 firescope and access all memory below 4GB using fireproxy from gdb.
1667 Even controlling a kernel debugger is possible using remote DMA.
1671 If ohci1394_dma=early is used as boot parameter, it will initialize
1672 all OHCI1394 controllers which are found in the PCI config space.
1674 As all changes to the FireWire bus such as enabling and disabling
1675 devices cause a bus reset and thereby disable remote DMA for all
1676 devices, be sure to have the cable plugged and FireWire enabled on
1677 the debugging host before booting the debug target for debugging.
1679 This code (~1k) is freed after boot. By then, the firewire stack
1680 in charge of the OHCI-1394 controllers should be used instead.
1682 See Documentation/debugging-via-ohci1394.txt for more information.
1684 menuconfig RUNTIME_TESTING_MENU
1685 bool "Runtime Testing"
1688 if RUNTIME_TESTING_MENU
1691 tristate "Linux Kernel Dump Test Tool Module"
1694 This module enables testing of the different dumping mechanisms by
1695 inducing system failures at predefined crash points.
1696 If you don't need it: say N
1697 Choose M here to compile this code as a module. The module will be
1700 Documentation on how to use the module can be found in
1701 Documentation/fault-injection/provoke-crashes.rst
1703 config TEST_LIST_SORT
1704 tristate "Linked list sorting test"
1705 depends on DEBUG_KERNEL || m
1707 Enable this to turn on 'list_sort()' function test. This test is
1708 executed only once during system boot (so affects only boot time),
1709 or at module load time.
1714 tristate "Array-based sort test"
1715 depends on DEBUG_KERNEL || m
1717 This option enables the self-test function of 'sort()' at boot,
1718 or at module load time.
1722 config KPROBES_SANITY_TEST
1723 bool "Kprobes sanity tests"
1724 depends on DEBUG_KERNEL
1727 This option provides for testing basic kprobes functionality on
1728 boot. Samples of kprobe and kretprobe are inserted and
1729 verified for functionality.
1731 Say N if you are unsure.
1733 config BACKTRACE_SELF_TEST
1734 tristate "Self test for the backtrace code"
1735 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 REED_SOLOMON_TEST
1755 tristate "Reed-Solomon library test"
1756 depends on DEBUG_KERNEL || m
1758 select REED_SOLOMON_ENC16
1759 select REED_SOLOMON_DEC16
1761 This option enables the self-test function of rslib at boot,
1762 or at module load time.
1766 config INTERVAL_TREE_TEST
1767 tristate "Interval tree test"
1768 depends on DEBUG_KERNEL
1769 select INTERVAL_TREE
1771 A benchmark measuring the performance of the interval tree library
1774 tristate "Per cpu operations test"
1775 depends on m && DEBUG_KERNEL
1777 Enable this option to build test module which validates per-cpu
1782 config ATOMIC64_SELFTEST
1783 tristate "Perform an atomic64_t self-test"
1785 Enable this option to test the atomic64_t functions at boot or
1786 at module load time.
1790 config ASYNC_RAID6_TEST
1791 tristate "Self test for hardware accelerated raid6 recovery"
1792 depends on ASYNC_RAID6_RECOV
1795 This is a one-shot self test that permutes through the
1796 recovery of all the possible two disk failure scenarios for a
1797 N-disk array. Recovery is performed with the asynchronous
1798 raid6 recovery routines, and will optionally use an offload
1799 engine if one is available.
1804 tristate "Test functions located in the hexdump module at runtime"
1806 config TEST_STRING_HELPERS
1807 tristate "Test functions located in the string_helpers module at runtime"
1810 tristate "Test strscpy*() family of functions at runtime"
1813 tristate "Test kstrto*() family of functions at runtime"
1816 tristate "Test printf() family of functions at runtime"
1819 tristate "Test bitmap_*() family of functions at runtime"
1821 Enable this option to test the bitmap functions at boot.
1825 config TEST_BITFIELD
1826 tristate "Test bitfield functions at runtime"
1828 Enable this option to test the bitfield functions at boot.
1833 tristate "Test functions located in the uuid module at runtime"
1836 tristate "Test the XArray code at runtime"
1838 config TEST_OVERFLOW
1839 tristate "Test check_*_overflow() functions at runtime"
1841 config TEST_RHASHTABLE
1842 tristate "Perform selftest on resizable hash table"
1844 Enable this option to test the rhashtable functions at boot.
1849 tristate "Perform selftest on hash functions"
1851 Enable this option to test the kernel's integer (<linux/hash.h>),
1852 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>)
1853 hash functions on boot (or module load).
1855 This is intended to help people writing architecture-specific
1856 optimized versions. If unsure, say N.
1859 tristate "Perform selftest on IDA functions"
1862 tristate "Perform selftest on priority array manager"
1865 Enable this option to test priority array manager on boot
1870 config TEST_IRQ_TIMINGS
1871 bool "IRQ timings selftest"
1872 depends on IRQ_TIMINGS
1874 Enable this option to test the irq timings code on boot.
1879 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
1892 tristate "Test module for stress/performance analysis of vmalloc allocator"
1897 This builds the "test_vmalloc" module that should be used for
1898 stress and performance analysis. So, any new change for vmalloc
1899 subsystem can be evaluated from performance and stability point
1904 config TEST_USER_COPY
1905 tristate "Test user/kernel boundary protections"
1908 This builds the "test_user_copy" module that runs sanity checks
1909 on the copy_to/from_user infrastructure, making sure basic
1910 user/kernel boundary testing is working. If it fails to load,
1911 a regression has been detected in the user/kernel memory boundary
1917 tristate "Test BPF filter functionality"
1920 This builds the "test_bpf" module that runs various test vectors
1921 against the BPF interpreter or BPF JIT compiler depending on the
1922 current setting. This is in particular useful for BPF JIT compiler
1923 development, but also to run regression tests against changes in
1924 the interpreter code. It also enables test stubs for eBPF maps and
1925 verifier used by user space verifier testsuite.
1929 config TEST_BLACKHOLE_DEV
1930 tristate "Test blackhole netdev functionality"
1933 This builds the "test_blackhole_dev" module that validates the
1934 data path through this blackhole netdev.
1938 config FIND_BIT_BENCHMARK
1939 tristate "Test find_bit functions"
1941 This builds the "test_find_bit" module that measure find_*_bit()
1942 functions performance.
1946 config TEST_FIRMWARE
1947 tristate "Test firmware loading via userspace interface"
1948 depends on FW_LOADER
1950 This builds the "test_firmware" module that creates a userspace
1951 interface for testing firmware loading. This can be used to
1952 control the triggering of firmware loading without needing an
1953 actual firmware-using device. The contents can be rechecked by
1959 tristate "sysctl test driver"
1960 depends on PROC_SYSCTL
1962 This builds the "test_sysctl" module. This driver enables to test the
1963 proc sysctl interfaces available to drivers safely without affecting
1964 production knobs which might alter system functionality.
1969 tristate "udelay test driver"
1971 This builds the "udelay_test" module that helps to make sure
1972 that udelay() is working properly.
1976 config TEST_STATIC_KEYS
1977 tristate "Test static keys"
1980 Test the static key interfaces.
1985 tristate "kmod stress tester"
1987 depends on NETDEVICES && NET_CORE && INET # for TUN
1994 Test the kernel's module loading mechanism: kmod. kmod implements
1995 support to load modules using the Linux kernel's usermode helper.
1996 This test provides a series of tests against kmod.
1998 Although technically you can either build test_kmod as a module or
1999 into the kernel we disallow building it into the kernel since
2000 it stress tests request_module() and this will very likely cause
2001 some issues by taking over precious threads available from other
2002 module load requests, ultimately this could be fatal.
2006 tools/testing/selftests/kmod/kmod.sh --help
2010 config TEST_DEBUG_VIRTUAL
2011 tristate "Test CONFIG_DEBUG_VIRTUAL feature"
2012 depends on DEBUG_VIRTUAL
2014 Test the kernel's ability to detect incorrect calls to
2015 virt_to_phys() done against the non-linear part of the
2016 kernel's virtual address map.
2020 config TEST_MEMCAT_P
2021 tristate "Test memcat_p() helper function"
2023 Test the memcat_p() helper for correctly merging two
2024 pointer arrays together.
2028 config TEST_LIVEPATCH
2029 tristate "Test livepatching"
2031 depends on DYNAMIC_DEBUG
2032 depends on LIVEPATCH
2035 Test kernel livepatching features for correctness. The tests will
2036 load test modules that will be livepatched in various scenarios.
2038 To run all the livepatching tests:
2040 make -C tools/testing/selftests TARGETS=livepatch run_tests
2042 Alternatively, individual tests may be invoked:
2044 tools/testing/selftests/livepatch/test-callbacks.sh
2045 tools/testing/selftests/livepatch/test-livepatch.sh
2046 tools/testing/selftests/livepatch/test-shadow-vars.sh
2051 tristate "Perform selftest on object aggreration manager"
2055 Enable this option to test object aggregation manager on boot
2059 config TEST_STACKINIT
2060 tristate "Test level of stack variable initialization"
2062 Test if the kernel is zero-initializing stack variables and
2063 padding. Coverage is controlled by compiler flags,
2064 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF,
2065 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL.
2070 tristate "Test heap/page initialization"
2072 Test if the kernel is zero-initializing heap and page allocations.
2073 This can be useful to test init_on_alloc and init_on_free features.
2077 endif # RUNTIME_TESTING_MENU
2082 This option adds a kernel parameter 'memtest', which allows memtest
2084 memtest=0, mean disabled; -- default
2085 memtest=1, mean do 1 test pattern;
2087 memtest=17, mean do 17 test patterns.
2088 If you are unsure how to answer this question, answer N.
2090 config BUG_ON_DATA_CORRUPTION
2091 bool "Trigger a BUG when data corruption is detected"
2094 Select this option if the kernel should BUG when it encounters
2095 data corruption in kernel memory structures when they get checked
2100 source "samples/Kconfig"
2102 source "lib/Kconfig.kgdb"
2104 source "lib/Kconfig.ubsan"
2106 config ARCH_HAS_DEVMEM_IS_ALLOWED
2109 config STRICT_DEVMEM
2110 bool "Filter access to /dev/mem"
2111 depends on MMU && DEVMEM
2112 depends on ARCH_HAS_DEVMEM_IS_ALLOWED
2113 default y if PPC || X86 || ARM64
2115 If this option is disabled, you allow userspace (root) access to all
2116 of memory, including kernel and userspace memory. Accidental
2117 access to this is obviously disastrous, but specific access can
2118 be used by people debugging the kernel. Note that with PAT support
2119 enabled, even in this case there are restrictions on /dev/mem
2120 use due to the cache aliasing requirements.
2122 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem
2123 file only allows userspace access to PCI space and the BIOS code and
2124 data regions. This is sufficient for dosemu and X and all common
2129 config IO_STRICT_DEVMEM
2130 bool "Filter I/O access to /dev/mem"
2131 depends on STRICT_DEVMEM
2133 If this option is disabled, you allow userspace (root) access to all
2134 io-memory regardless of whether a driver is actively using that
2135 range. Accidental access to this is obviously disastrous, but
2136 specific access can be used by people debugging kernel drivers.
2138 If this option is switched on, the /dev/mem file only allows
2139 userspace access to *idle* io-memory ranges (see /proc/iomem) This
2140 may break traditional users of /dev/mem (dosemu, legacy X, etc...)
2141 if the driver using a given range cannot be disabled.
2145 source "arch/$(SRCARCH)/Kconfig.debug"
2147 endmenu # Kernel hacking
projects / linux-2.6-microblaze.git / blob