1menu "printk and dmesg options" 2 3config PRINTK_TIME 4 bool "Show timing information on printks" 5 depends on PRINTK 6 help 7 Selecting this option causes time stamps of the printk() 8 messages to be added to the output of the syslog() system 9 call and at the console. 10 11 The timestamp is always recorded internally, and exported 12 to /dev/kmsg. This flag just specifies if the timestamp should 13 be included, not that the timestamp is recorded. 14 15 The behavior is also controlled by the kernel command line 16 parameter printk.time=1. See Documentation/kernel-parameters.txt 17 18config MESSAGE_LOGLEVEL_DEFAULT 19 int "Default message log level (1-7)" 20 range 1 7 21 default "4" 22 help 23 Default log level for printk statements with no specified priority. 24 25 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks 26 that are auditing their logs closely may want to set it to a lower 27 priority. 28 29config BOOT_PRINTK_DELAY 30 bool "Delay each boot printk message by N milliseconds" 31 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY 32 help 33 This build option allows you to read kernel boot messages 34 by inserting a short delay after each one. The delay is 35 specified in milliseconds on the kernel command line, 36 using "boot_delay=N". 37 38 It is likely that you would also need to use "lpj=M" to preset 39 the "loops per jiffie" value. 40 See a previous boot log for the "lpj" value to use for your 41 system, and then set "lpj=M" before setting "boot_delay=N". 42 NOTE: Using this option may adversely affect SMP systems. 43 I.e., processors other than the first one may not boot up. 44 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect 45 what it believes to be lockup conditions. 46 47config DYNAMIC_DEBUG 48 bool "Enable dynamic printk() support" 49 default n 50 depends on PRINTK 51 depends on DEBUG_FS 52 help 53 54 Compiles debug level messages into the kernel, which would not 55 otherwise be available at runtime. These messages can then be 56 enabled/disabled based on various levels of scope - per source file, 57 function, module, format string, and line number. This mechanism 58 implicitly compiles in all pr_debug() and dev_dbg() calls, which 59 enlarges the kernel text size by about 2%. 60 61 If a source file is compiled with DEBUG flag set, any 62 pr_debug() calls in it are enabled by default, but can be 63 disabled at runtime as below. Note that DEBUG flag is 64 turned on by many CONFIG_*DEBUG* options. 65 66 Usage: 67 68 Dynamic debugging is controlled via the 'dynamic_debug/control' file, 69 which is contained in the 'debugfs' filesystem. Thus, the debugfs 70 filesystem must first be mounted before making use of this feature. 71 We refer the control file as: <debugfs>/dynamic_debug/control. This 72 file contains a list of the debug statements that can be enabled. The 73 format for each line of the file is: 74 75 filename:lineno [module]function flags format 76 77 filename : source file of the debug statement 78 lineno : line number of the debug statement 79 module : module that contains the debug statement 80 function : function that contains the debug statement 81 flags : '=p' means the line is turned 'on' for printing 82 format : the format used for the debug statement 83 84 From a live system: 85 86 nullarbor:~ # cat <debugfs>/dynamic_debug/control 87 # filename:lineno [module]function flags format 88 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012" 89 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012" 90 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012" 91 92 Example usage: 93 94 // enable the message at line 1603 of file svcsock.c 95 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > 96 <debugfs>/dynamic_debug/control 97 98 // enable all the messages in file svcsock.c 99 nullarbor:~ # echo -n 'file svcsock.c +p' > 100 <debugfs>/dynamic_debug/control 101 102 // enable all the messages in the NFS server module 103 nullarbor:~ # echo -n 'module nfsd +p' > 104 <debugfs>/dynamic_debug/control 105 106 // enable all 12 messages in the function svc_process() 107 nullarbor:~ # echo -n 'func svc_process +p' > 108 <debugfs>/dynamic_debug/control 109 110 // disable all 12 messages in the function svc_process() 111 nullarbor:~ # echo -n 'func svc_process -p' > 112 <debugfs>/dynamic_debug/control 113 114 See Documentation/dynamic-debug-howto.txt for additional information. 115 116endmenu # "printk and dmesg options" 117 118menu "Compile-time checks and compiler options" 119 120config DEBUG_INFO 121 bool "Compile the kernel with debug info" 122 depends on DEBUG_KERNEL && !COMPILE_TEST 123 help 124 If you say Y here the resulting kernel image will include 125 debugging info resulting in a larger kernel image. 126 This adds debug symbols to the kernel and modules (gcc -g), and 127 is needed if you intend to use kernel crashdump or binary object 128 tools like crash, kgdb, LKCD, gdb, etc on the kernel. 129 Say Y here only if you plan to debug the kernel. 130 131 If unsure, say N. 132 133config DEBUG_INFO_REDUCED 134 bool "Reduce debugging information" 135 depends on DEBUG_INFO 136 help 137 If you say Y here gcc is instructed to generate less debugging 138 information for structure types. This means that tools that 139 need full debugging information (like kgdb or systemtap) won't 140 be happy. But if you merely need debugging information to 141 resolve line numbers there is no loss. Advantage is that 142 build directory object sizes shrink dramatically over a full 143 DEBUG_INFO build and compile times are reduced too. 144 Only works with newer gcc versions. 145 146config DEBUG_INFO_SPLIT 147 bool "Produce split debuginfo in .dwo files" 148 depends on DEBUG_INFO 149 help 150 Generate debug info into separate .dwo files. This significantly 151 reduces the build directory size for builds with DEBUG_INFO, 152 because it stores the information only once on disk in .dwo 153 files instead of multiple times in object files and executables. 154 In addition the debug information is also compressed. 155 156 Requires recent gcc (4.7+) and recent gdb/binutils. 157 Any tool that packages or reads debug information would need 158 to know about the .dwo files and include them. 159 Incompatible with older versions of ccache. 160 161config DEBUG_INFO_DWARF4 162 bool "Generate dwarf4 debuginfo" 163 depends on DEBUG_INFO 164 help 165 Generate dwarf4 debug info. This requires recent versions 166 of gcc and gdb. It makes the debug information larger. 167 But it significantly improves the success of resolving 168 variables in gdb on optimized code. 169 170config ENABLE_WARN_DEPRECATED 171 bool "Enable __deprecated logic" 172 default y 173 help 174 Enable the __deprecated logic in the kernel build. 175 Disable this to suppress the "warning: 'foo' is deprecated 176 (declared at kernel/power/somefile.c:1234)" messages. 177 178config ENABLE_MUST_CHECK 179 bool "Enable __must_check logic" 180 default y 181 help 182 Enable the __must_check logic in the kernel build. Disable this to 183 suppress the "warning: ignoring return value of 'foo', declared with 184 attribute warn_unused_result" messages. 185 186config FRAME_WARN 187 int "Warn for stack frames larger than (needs gcc 4.4)" 188 range 0 8192 189 default 1024 if !64BIT 190 default 2048 if 64BIT 191 help 192 Tell gcc to warn at build time for stack frames larger than this. 193 Setting this too low will cause a lot of warnings. 194 Setting it to 0 disables the warning. 195 Requires gcc 4.4 196 197config STRIP_ASM_SYMS 198 bool "Strip assembler-generated symbols during link" 199 default n 200 help 201 Strip internal assembler-generated symbols during a link (symbols 202 that look like '.Lxxx') so they don't pollute the output of 203 get_wchan() and suchlike. 204 205config READABLE_ASM 206 bool "Generate readable assembler code" 207 depends on DEBUG_KERNEL 208 help 209 Disable some compiler optimizations that tend to generate human unreadable 210 assembler output. This may make the kernel slightly slower, but it helps 211 to keep kernel developers who have to stare a lot at assembler listings 212 sane. 213 214config UNUSED_SYMBOLS 215 bool "Enable unused/obsolete exported symbols" 216 default y if X86 217 help 218 Unused but exported symbols make the kernel needlessly bigger. For 219 that reason most of these unused exports will soon be removed. This 220 option is provided temporarily to provide a transition period in case 221 some external kernel module needs one of these symbols anyway. If you 222 encounter such a case in your module, consider if you are actually 223 using the right API. (rationale: since nobody in the kernel is using 224 this in a module, there is a pretty good chance it's actually the 225 wrong interface to use). If you really need the symbol, please send a 226 mail to the linux kernel mailing list mentioning the symbol and why 227 you really need it, and what the merge plan to the mainline kernel for 228 your module is. 229 230config DEBUG_FS 231 bool "Debug Filesystem" 232 help 233 debugfs is a virtual file system that kernel developers use to put 234 debugging files into. Enable this option to be able to read and 235 write to these files. 236 237 For detailed documentation on the debugfs API, see 238 Documentation/DocBook/filesystems. 239 240 If unsure, say N. 241 242config HEADERS_CHECK 243 bool "Run 'make headers_check' when building vmlinux" 244 depends on !UML 245 help 246 This option will extract the user-visible kernel headers whenever 247 building the kernel, and will run basic sanity checks on them to 248 ensure that exported files do not attempt to include files which 249 were not exported, etc. 250 251 If you're making modifications to header files which are 252 relevant for userspace, say 'Y', and check the headers 253 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in 254 your build tree), to make sure they're suitable. 255 256config DEBUG_SECTION_MISMATCH 257 bool "Enable full Section mismatch analysis" 258 help 259 The section mismatch analysis checks if there are illegal 260 references from one section to another section. 261 During linktime or runtime, some sections are dropped; 262 any use of code/data previously in these sections would 263 most likely result in an oops. 264 In the code, functions and variables are annotated with 265 __init,, etc. (see the full list in include/linux/init.h), 266 which results in the code/data being placed in specific sections. 267 The section mismatch analysis is always performed after a full 268 kernel build, and enabling this option causes the following 269 additional steps to occur: 270 - Add the option -fno-inline-functions-called-once to gcc commands. 271 When inlining a function annotated with __init in a non-init 272 function, we would lose the section information and thus 273 the analysis would not catch the illegal reference. 274 This option tells gcc to inline less (but it does result in 275 a larger kernel). 276 - Run the section mismatch analysis for each module/built-in.o file. 277 When we run the section mismatch analysis on vmlinux.o, we 278 lose valueble information about where the mismatch was 279 introduced. 280 Running the analysis for each module/built-in.o file 281 tells where the mismatch happens much closer to the 282 source. The drawback is that the same mismatch is 283 reported at least twice. 284 - Enable verbose reporting from modpost in order to help resolve 285 the section mismatches that are reported. 286 287# 288# Select this config option from the architecture Kconfig, if it 289# is preferred to always offer frame pointers as a config 290# option on the architecture (regardless of KERNEL_DEBUG): 291# 292config ARCH_WANT_FRAME_POINTERS 293 bool 294 help 295 296config FRAME_POINTER 297 bool "Compile the kernel with frame pointers" 298 depends on DEBUG_KERNEL && \ 299 (CRIS || M68K || FRV || UML || \ 300 AVR32 || SUPERH || BLACKFIN || MN10300 || METAG) || \ 301 ARCH_WANT_FRAME_POINTERS 302 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 303 help 304 If you say Y here the resulting kernel image will be slightly 305 larger and slower, but it gives very useful debugging information 306 in case of kernel bugs. (precise oopses/stacktraces/warnings) 307 308config DEBUG_FORCE_WEAK_PER_CPU 309 bool "Force weak per-cpu definitions" 310 depends on DEBUG_KERNEL 311 help 312 s390 and alpha require percpu variables in modules to be 313 defined weak to work around addressing range issue which 314 puts the following two restrictions on percpu variable 315 definitions. 316 317 1. percpu symbols must be unique whether static or not 318 2. percpu variables can't be defined inside a function 319 320 To ensure that generic code follows the above rules, this 321 option forces all percpu variables to be defined as weak. 322 323endmenu # "Compiler options" 324 325config MAGIC_SYSRQ 326 bool "Magic SysRq key" 327 depends on !UML 328 help 329 If you say Y here, you will have some control over the system even 330 if the system crashes for example during kernel debugging (e.g., you 331 will be able to flush the buffer cache to disk, reboot the system 332 immediately or dump some status information). This is accomplished 333 by pressing various keys while holding SysRq (Alt+PrintScreen). It 334 also works on a serial console (on PC hardware at least), if you 335 send a BREAK and then within 5 seconds a command keypress. The 336 keys are documented in <file:Documentation/sysrq.txt>. Don't say Y 337 unless you really know what this hack does. 338 339config MAGIC_SYSRQ_DEFAULT_ENABLE 340 hex "Enable magic SysRq key functions by default" 341 depends on MAGIC_SYSRQ 342 default 0x1 343 help 344 Specifies which SysRq key functions are enabled by default. 345 This may be set to 1 or 0 to enable or disable them all, or 346 to a bitmask as described in Documentation/sysrq.txt. 347 348config DEBUG_KERNEL 349 bool "Kernel debugging" 350 help 351 Say Y here if you are developing drivers or trying to debug and 352 identify kernel problems. 353 354menu "Memory Debugging" 355 356source mm/Kconfig.debug 357 358config DEBUG_OBJECTS 359 bool "Debug object operations" 360 depends on DEBUG_KERNEL 361 help 362 If you say Y here, additional code will be inserted into the 363 kernel to track the life time of various objects and validate 364 the operations on those objects. 365 366config DEBUG_OBJECTS_SELFTEST 367 bool "Debug objects selftest" 368 depends on DEBUG_OBJECTS 369 help 370 This enables the selftest of the object debug code. 371 372config DEBUG_OBJECTS_FREE 373 bool "Debug objects in freed memory" 374 depends on DEBUG_OBJECTS 375 help 376 This enables checks whether a k/v free operation frees an area 377 which contains an object which has not been deactivated 378 properly. This can make kmalloc/kfree-intensive workloads 379 much slower. 380 381config DEBUG_OBJECTS_TIMERS 382 bool "Debug timer objects" 383 depends on DEBUG_OBJECTS 384 help 385 If you say Y here, additional code will be inserted into the 386 timer routines to track the life time of timer objects and 387 validate the timer operations. 388 389config DEBUG_OBJECTS_WORK 390 bool "Debug work objects" 391 depends on DEBUG_OBJECTS 392 help 393 If you say Y here, additional code will be inserted into the 394 work queue routines to track the life time of work objects and 395 validate the work operations. 396 397config DEBUG_OBJECTS_RCU_HEAD 398 bool "Debug RCU callbacks objects" 399 depends on DEBUG_OBJECTS 400 help 401 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 402 403config DEBUG_OBJECTS_PERCPU_COUNTER 404 bool "Debug percpu counter objects" 405 depends on DEBUG_OBJECTS 406 help 407 If you say Y here, additional code will be inserted into the 408 percpu counter routines to track the life time of percpu counter 409 objects and validate the percpu counter operations. 410 411config DEBUG_OBJECTS_ENABLE_DEFAULT 412 int "debug_objects bootup default value (0-1)" 413 range 0 1 414 default "1" 415 depends on DEBUG_OBJECTS 416 help 417 Debug objects boot parameter default value 418 419config DEBUG_SLAB 420 bool "Debug slab memory allocations" 421 depends on DEBUG_KERNEL && SLAB && !KMEMCHECK 422 help 423 Say Y here to have the kernel do limited verification on memory 424 allocation as well as poisoning memory on free to catch use of freed 425 memory. This can make kmalloc/kfree-intensive workloads much slower. 426 427config DEBUG_SLAB_LEAK 428 bool "Memory leak debugging" 429 depends on DEBUG_SLAB 430 431config SLUB_DEBUG_ON 432 bool "SLUB debugging on by default" 433 depends on SLUB && SLUB_DEBUG && !KMEMCHECK 434 default n 435 help 436 Boot with debugging on by default. SLUB boots by default with 437 the runtime debug capabilities switched off. Enabling this is 438 equivalent to specifying the "slub_debug" parameter on boot. 439 There is no support for more fine grained debug control like 440 possible with slub_debug=xxx. SLUB debugging may be switched 441 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 442 "slub_debug=-". 443 444config SLUB_STATS 445 default n 446 bool "Enable SLUB performance statistics" 447 depends on SLUB && SYSFS 448 help 449 SLUB statistics are useful to debug SLUBs allocation behavior in 450 order find ways to optimize the allocator. This should never be 451 enabled for production use since keeping statistics slows down 452 the allocator by a few percentage points. The slabinfo command 453 supports the determination of the most active slabs to figure 454 out which slabs are relevant to a particular load. 455 Try running: slabinfo -DA 456 457config HAVE_DEBUG_KMEMLEAK 458 bool 459 460config DEBUG_KMEMLEAK 461 bool "Kernel memory leak detector" 462 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 463 select DEBUG_FS 464 select STACKTRACE if STACKTRACE_SUPPORT 465 select KALLSYMS 466 select CRC32 467 help 468 Say Y here if you want to enable the memory leak 469 detector. The memory allocation/freeing is traced in a way 470 similar to the Boehm's conservative garbage collector, the 471 difference being that the orphan objects are not freed but 472 only shown in /sys/kernel/debug/kmemleak. Enabling this 473 feature will introduce an overhead to memory 474 allocations. See Documentation/kmemleak.txt for more 475 details. 476 477 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 478 of finding leaks due to the slab objects poisoning. 479 480 In order to access the kmemleak file, debugfs needs to be 481 mounted (usually at /sys/kernel/debug). 482 483config DEBUG_KMEMLEAK_EARLY_LOG_SIZE 484 int "Maximum kmemleak early log entries" 485 depends on DEBUG_KMEMLEAK 486 range 200 40000 487 default 400 488 help 489 Kmemleak must track all the memory allocations to avoid 490 reporting false positives. Since memory may be allocated or 491 freed before kmemleak is initialised, an early log buffer is 492 used to store these actions. If kmemleak reports "early log 493 buffer exceeded", please increase this value. 494 495config DEBUG_KMEMLEAK_TEST 496 tristate "Simple test for the kernel memory leak detector" 497 depends on DEBUG_KMEMLEAK && m 498 help 499 This option enables a module that explicitly leaks memory. 500 501 If unsure, say N. 502 503config DEBUG_KMEMLEAK_DEFAULT_OFF 504 bool "Default kmemleak to off" 505 depends on DEBUG_KMEMLEAK 506 help 507 Say Y here to disable kmemleak by default. It can then be enabled 508 on the command line via kmemleak=on. 509 510config DEBUG_STACK_USAGE 511 bool "Stack utilization instrumentation" 512 depends on DEBUG_KERNEL && !IA64 && !PARISC && !METAG 513 help 514 Enables the display of the minimum amount of free stack which each 515 task has ever had available in the sysrq-T and sysrq-P debug output. 516 517 This option will slow down process creation somewhat. 518 519config DEBUG_VM 520 bool "Debug VM" 521 depends on DEBUG_KERNEL 522 help 523 Enable this to turn on extended checks in the virtual-memory system 524 that may impact performance. 525 526 If unsure, say N. 527 528config DEBUG_VM_VMACACHE 529 bool "Debug VMA caching" 530 depends on DEBUG_VM 531 help 532 Enable this to turn on VMA caching debug information. Doing so 533 can cause significant overhead, so only enable it in non-production 534 environments. 535 536 If unsure, say N. 537 538config DEBUG_VM_RB 539 bool "Debug VM red-black trees" 540 depends on DEBUG_VM 541 help 542 Enable VM red-black tree debugging information and extra validations. 543 544 If unsure, say N. 545 546config DEBUG_VIRTUAL 547 bool "Debug VM translations" 548 depends on DEBUG_KERNEL && X86 549 help 550 Enable some costly sanity checks in virtual to page code. This can 551 catch mistakes with virt_to_page() and friends. 552 553 If unsure, say N. 554 555config DEBUG_NOMMU_REGIONS 556 bool "Debug the global anon/private NOMMU mapping region tree" 557 depends on DEBUG_KERNEL && !MMU 558 help 559 This option causes the global tree of anonymous and private mapping 560 regions to be regularly checked for invalid topology. 561 562config DEBUG_MEMORY_INIT 563 bool "Debug memory initialisation" if EXPERT 564 default !EXPERT 565 help 566 Enable this for additional checks during memory initialisation. 567 The sanity checks verify aspects of the VM such as the memory model 568 and other information provided by the architecture. Verbose 569 information will be printed at KERN_DEBUG loglevel depending 570 on the mminit_loglevel= command-line option. 571 572 If unsure, say Y 573 574config MEMORY_NOTIFIER_ERROR_INJECT 575 tristate "Memory hotplug notifier error injection module" 576 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 577 help 578 This option provides the ability to inject artificial errors to 579 memory hotplug notifier chain callbacks. It is controlled through 580 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 581 582 If the notifier call chain should be failed with some events 583 notified, write the error code to "actions/<notifier event>/error". 584 585 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 586 587 # cd /sys/kernel/debug/notifier-error-inject/memory 588 # echo -12 > actions/MEM_GOING_OFFLINE/error 589 # echo offline > /sys/devices/system/memory/memoryXXX/state 590 bash: echo: write error: Cannot allocate memory 591 592 To compile this code as a module, choose M here: the module will 593 be called memory-notifier-error-inject. 594 595 If unsure, say N. 596 597config DEBUG_PER_CPU_MAPS 598 bool "Debug access to per_cpu maps" 599 depends on DEBUG_KERNEL 600 depends on SMP 601 help 602 Say Y to verify that the per_cpu map being accessed has 603 been set up. This adds a fair amount of code to kernel memory 604 and decreases performance. 605 606 Say N if unsure. 607 608config DEBUG_HIGHMEM 609 bool "Highmem debugging" 610 depends on DEBUG_KERNEL && HIGHMEM 611 help 612 This option enables additional error checking for high memory 613 systems. Disable for production systems. 614 615config HAVE_DEBUG_STACKOVERFLOW 616 bool 617 618config DEBUG_STACKOVERFLOW 619 bool "Check for stack overflows" 620 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 621 ---help--- 622 Say Y here if you want to check for overflows of kernel, IRQ 623 and exception stacks (if your archicture uses them). This 624 option will show detailed messages if free stack space drops 625 below a certain limit. 626 627 These kinds of bugs usually occur when call-chains in the 628 kernel get too deep, especially when interrupts are 629 involved. 630 631 Use this in cases where you see apparently random memory 632 corruption, especially if it appears in 'struct thread_info' 633 634 If in doubt, say "N". 635 636source "lib/Kconfig.kmemcheck" 637 638endmenu # "Memory Debugging" 639 640config DEBUG_SHIRQ 641 bool "Debug shared IRQ handlers" 642 depends on DEBUG_KERNEL 643 help 644 Enable this to generate a spurious interrupt as soon as a shared 645 interrupt handler is registered, and just before one is deregistered. 646 Drivers ought to be able to handle interrupts coming in at those 647 points; some don't and need to be caught. 648 649menu "Debug Lockups and Hangs" 650 651config LOCKUP_DETECTOR 652 bool "Detect Hard and Soft Lockups" 653 depends on DEBUG_KERNEL && !S390 654 help 655 Say Y here to enable the kernel to act as a watchdog to detect 656 hard and soft lockups. 657 658 Softlockups are bugs that cause the kernel to loop in kernel 659 mode for more than 20 seconds, without giving other tasks a 660 chance to run. The current stack trace is displayed upon 661 detection and the system will stay locked up. 662 663 Hardlockups are bugs that cause the CPU to loop in kernel mode 664 for more than 10 seconds, without letting other interrupts have a 665 chance to run. The current stack trace is displayed upon detection 666 and the system will stay locked up. 667 668 The overhead should be minimal. A periodic hrtimer runs to 669 generate interrupts and kick the watchdog task every 4 seconds. 670 An NMI is generated every 10 seconds or so to check for hardlockups. 671 672 The frequency of hrtimer and NMI events and the soft and hard lockup 673 thresholds can be controlled through the sysctl watchdog_thresh. 674 675config HARDLOCKUP_DETECTOR 676 def_bool y 677 depends on LOCKUP_DETECTOR && !HAVE_NMI_WATCHDOG 678 depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI 679 680config BOOTPARAM_HARDLOCKUP_PANIC 681 bool "Panic (Reboot) On Hard Lockups" 682 depends on HARDLOCKUP_DETECTOR 683 help 684 Say Y here to enable the kernel to panic on "hard lockups", 685 which are bugs that cause the kernel to loop in kernel 686 mode with interrupts disabled for more than 10 seconds (configurable 687 using the watchdog_thresh sysctl). 688 689 Say N if unsure. 690 691config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 692 int 693 depends on HARDLOCKUP_DETECTOR 694 range 0 1 695 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 696 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 697 698config BOOTPARAM_SOFTLOCKUP_PANIC 699 bool "Panic (Reboot) On Soft Lockups" 700 depends on LOCKUP_DETECTOR 701 help 702 Say Y here to enable the kernel to panic on "soft lockups", 703 which are bugs that cause the kernel to loop in kernel 704 mode for more than 20 seconds (configurable using the watchdog_thresh 705 sysctl), without giving other tasks a chance to run. 706 707 The panic can be used in combination with panic_timeout, 708 to cause the system to reboot automatically after a 709 lockup has been detected. This feature is useful for 710 high-availability systems that have uptime guarantees and 711 where a lockup must be resolved ASAP. 712 713 Say N if unsure. 714 715config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 716 int 717 depends on LOCKUP_DETECTOR 718 range 0 1 719 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 720 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 721 722config DETECT_HUNG_TASK 723 bool "Detect Hung Tasks" 724 depends on DEBUG_KERNEL 725 default LOCKUP_DETECTOR 726 help 727 Say Y here to enable the kernel to detect "hung tasks", 728 which are bugs that cause the task to be stuck in 729 uninterruptible "D" state indefinitiley. 730 731 When a hung task is detected, the kernel will print the 732 current stack trace (which you should report), but the 733 task will stay in uninterruptible state. If lockdep is 734 enabled then all held locks will also be reported. This 735 feature has negligible overhead. 736 737config DEFAULT_HUNG_TASK_TIMEOUT 738 int "Default timeout for hung task detection (in seconds)" 739 depends on DETECT_HUNG_TASK 740 default 120 741 help 742 This option controls the default timeout (in seconds) used 743 to determine when a task has become non-responsive and should 744 be considered hung. 745 746 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 747 sysctl or by writing a value to 748 /proc/sys/kernel/hung_task_timeout_secs. 749 750 A timeout of 0 disables the check. The default is two minutes. 751 Keeping the default should be fine in most cases. 752 753config BOOTPARAM_HUNG_TASK_PANIC 754 bool "Panic (Reboot) On Hung Tasks" 755 depends on DETECT_HUNG_TASK 756 help 757 Say Y here to enable the kernel to panic on "hung tasks", 758 which are bugs that cause the kernel to leave a task stuck 759 in uninterruptible "D" state. 760 761 The panic can be used in combination with panic_timeout, 762 to cause the system to reboot automatically after a 763 hung task has been detected. This feature is useful for 764 high-availability systems that have uptime guarantees and 765 where a hung tasks must be resolved ASAP. 766 767 Say N if unsure. 768 769config BOOTPARAM_HUNG_TASK_PANIC_VALUE 770 int 771 depends on DETECT_HUNG_TASK 772 range 0 1 773 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 774 default 1 if BOOTPARAM_HUNG_TASK_PANIC 775 776endmenu # "Debug lockups and hangs" 777 778config PANIC_ON_OOPS 779 bool "Panic on Oops" 780 help 781 Say Y here to enable the kernel to panic when it oopses. This 782 has the same effect as setting oops=panic on the kernel command 783 line. 784 785 This feature is useful to ensure that the kernel does not do 786 anything erroneous after an oops which could result in data 787 corruption or other issues. 788 789 Say N if unsure. 790 791config PANIC_ON_OOPS_VALUE 792 int 793 range 0 1 794 default 0 if !PANIC_ON_OOPS 795 default 1 if PANIC_ON_OOPS 796 797config PANIC_TIMEOUT 798 int "panic timeout" 799 default 0 800 help 801 Set the timeout value (in seconds) until a reboot occurs when the 802 the kernel panics. If n = 0, then we wait forever. A timeout 803 value n > 0 will wait n seconds before rebooting, while a timeout 804 value n < 0 will reboot immediately. 805 806config SCHED_DEBUG 807 bool "Collect scheduler debugging info" 808 depends on DEBUG_KERNEL && PROC_FS 809 default y 810 help 811 If you say Y here, the /proc/sched_debug file will be provided 812 that can help debug the scheduler. The runtime overhead of this 813 option is minimal. 814 815config SCHEDSTATS 816 bool "Collect scheduler statistics" 817 depends on DEBUG_KERNEL && PROC_FS 818 help 819 If you say Y here, additional code will be inserted into the 820 scheduler and related routines to collect statistics about 821 scheduler behavior and provide them in /proc/schedstat. These 822 stats may be useful for both tuning and debugging the scheduler 823 If you aren't debugging the scheduler or trying to tune a specific 824 application, you can say N to avoid the very slight overhead 825 this adds. 826 827config TIMER_STATS 828 bool "Collect kernel timers statistics" 829 depends on DEBUG_KERNEL && PROC_FS 830 help 831 If you say Y here, additional code will be inserted into the 832 timer routines to collect statistics about kernel timers being 833 reprogrammed. The statistics can be read from /proc/timer_stats. 834 The statistics collection is started by writing 1 to /proc/timer_stats, 835 writing 0 stops it. This feature is useful to collect information 836 about timer usage patterns in kernel and userspace. This feature 837 is lightweight if enabled in the kernel config but not activated 838 (it defaults to deactivated on bootup and will only be activated 839 if some application like powertop activates it explicitly). 840 841config DEBUG_PREEMPT 842 bool "Debug preemptible kernel" 843 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT 844 default y 845 help 846 If you say Y here then the kernel will use a debug variant of the 847 commonly used smp_processor_id() function and will print warnings 848 if kernel code uses it in a preemption-unsafe way. Also, the kernel 849 will detect preemption count underflows. 850 851menu "Lock Debugging (spinlocks, mutexes, etc...)" 852 853config DEBUG_RT_MUTEXES 854 bool "RT Mutex debugging, deadlock detection" 855 depends on DEBUG_KERNEL && RT_MUTEXES 856 help 857 This allows rt mutex semantics violations and rt mutex related 858 deadlocks (lockups) to be detected and reported automatically. 859 860config RT_MUTEX_TESTER 861 bool "Built-in scriptable tester for rt-mutexes" 862 depends on DEBUG_KERNEL && RT_MUTEXES && BROKEN 863 help 864 This option enables a rt-mutex tester. 865 866config DEBUG_SPINLOCK 867 bool "Spinlock and rw-lock debugging: basic checks" 868 depends on DEBUG_KERNEL 869 select UNINLINE_SPIN_UNLOCK 870 help 871 Say Y here and build SMP to catch missing spinlock initialization 872 and certain other kinds of spinlock errors commonly made. This is 873 best used in conjunction with the NMI watchdog so that spinlock 874 deadlocks are also debuggable. 875 876config DEBUG_MUTEXES 877 bool "Mutex debugging: basic checks" 878 depends on DEBUG_KERNEL 879 help 880 This feature allows mutex semantics violations to be detected and 881 reported. 882 883config DEBUG_WW_MUTEX_SLOWPATH 884 bool "Wait/wound mutex debugging: Slowpath testing" 885 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 886 select DEBUG_LOCK_ALLOC 887 select DEBUG_SPINLOCK 888 select DEBUG_MUTEXES 889 help 890 This feature enables slowpath testing for w/w mutex users by 891 injecting additional -EDEADLK wound/backoff cases. Together with 892 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 893 will test all possible w/w mutex interface abuse with the 894 exception of simply not acquiring all the required locks. 895 896config DEBUG_LOCK_ALLOC 897 bool "Lock debugging: detect incorrect freeing of live locks" 898 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 899 select DEBUG_SPINLOCK 900 select DEBUG_MUTEXES 901 select LOCKDEP 902 help 903 This feature will check whether any held lock (spinlock, rwlock, 904 mutex or rwsem) is incorrectly freed by the kernel, via any of the 905 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 906 vfree(), etc.), whether a live lock is incorrectly reinitialized via 907 spin_lock_init()/mutex_init()/etc., or whether there is any lock 908 held during task exit. 909 910config PROVE_LOCKING 911 bool "Lock debugging: prove locking correctness" 912 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 913 select LOCKDEP 914 select DEBUG_SPINLOCK 915 select DEBUG_MUTEXES 916 select DEBUG_LOCK_ALLOC 917 select TRACE_IRQFLAGS 918 default n 919 help 920 This feature enables the kernel to prove that all locking 921 that occurs in the kernel runtime is mathematically 922 correct: that under no circumstance could an arbitrary (and 923 not yet triggered) combination of observed locking 924 sequences (on an arbitrary number of CPUs, running an 925 arbitrary number of tasks and interrupt contexts) cause a 926 deadlock. 927 928 In short, this feature enables the kernel to report locking 929 related deadlocks before they actually occur. 930 931 The proof does not depend on how hard and complex a 932 deadlock scenario would be to trigger: how many 933 participant CPUs, tasks and irq-contexts would be needed 934 for it to trigger. The proof also does not depend on 935 timing: if a race and a resulting deadlock is possible 936 theoretically (no matter how unlikely the race scenario 937 is), it will be proven so and will immediately be 938 reported by the kernel (once the event is observed that 939 makes the deadlock theoretically possible). 940 941 If a deadlock is impossible (i.e. the locking rules, as 942 observed by the kernel, are mathematically correct), the 943 kernel reports nothing. 944 945 NOTE: this feature can also be enabled for rwlocks, mutexes 946 and rwsems - in which case all dependencies between these 947 different locking variants are observed and mapped too, and 948 the proof of observed correctness is also maintained for an 949 arbitrary combination of these separate locking variants. 950 951 For more details, see Documentation/lockdep-design.txt. 952 953config LOCKDEP 954 bool 955 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 956 select STACKTRACE 957 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE && !ARC && !SCORE 958 select KALLSYMS 959 select KALLSYMS_ALL 960 961config LOCK_STAT 962 bool "Lock usage statistics" 963 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 964 select LOCKDEP 965 select DEBUG_SPINLOCK 966 select DEBUG_MUTEXES 967 select DEBUG_LOCK_ALLOC 968 default n 969 help 970 This feature enables tracking lock contention points 971 972 For more details, see Documentation/lockstat.txt 973 974 This also enables lock events required by "perf lock", 975 subcommand of perf. 976 If you want to use "perf lock", you also need to turn on 977 CONFIG_EVENT_TRACING. 978 979 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 980 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 981 982config DEBUG_LOCKDEP 983 bool "Lock dependency engine debugging" 984 depends on DEBUG_KERNEL && LOCKDEP 985 help 986 If you say Y here, the lock dependency engine will do 987 additional runtime checks to debug itself, at the price 988 of more runtime overhead. 989 990config DEBUG_ATOMIC_SLEEP 991 bool "Sleep inside atomic section checking" 992 select PREEMPT_COUNT 993 depends on DEBUG_KERNEL 994 help 995 If you say Y here, various routines which may sleep will become very 996 noisy if they are called inside atomic sections: when a spinlock is 997 held, inside an rcu read side critical section, inside preempt disabled 998 sections, inside an interrupt, etc... 999 1000config DEBUG_LOCKING_API_SELFTESTS 1001 bool "Locking API boot-time self-tests" 1002 depends on DEBUG_KERNEL 1003 help 1004 Say Y here if you want the kernel to run a short self-test during 1005 bootup. The self-test checks whether common types of locking bugs 1006 are detected by debugging mechanisms or not. (if you disable 1007 lock debugging then those bugs wont be detected of course.) 1008 The following locking APIs are covered: spinlocks, rwlocks, 1009 mutexes and rwsems. 1010 1011config LOCK_TORTURE_TEST 1012 tristate "torture tests for locking" 1013 depends on DEBUG_KERNEL 1014 select TORTURE_TEST 1015 default n 1016 help 1017 This option provides a kernel module that runs torture tests 1018 on kernel locking primitives. The kernel module may be built 1019 after the fact on the running kernel to be tested, if desired. 1020 1021 Say Y here if you want kernel locking-primitive torture tests 1022 to be built into the kernel. 1023 Say M if you want these torture tests to build as a module. 1024 Say N if you are unsure. 1025 1026endmenu # lock debugging 1027 1028config TRACE_IRQFLAGS 1029 bool 1030 help 1031 Enables hooks to interrupt enabling and disabling for 1032 either tracing or lock debugging. 1033 1034config STACKTRACE 1035 bool 1036 depends on STACKTRACE_SUPPORT 1037 1038config DEBUG_KOBJECT 1039 bool "kobject debugging" 1040 depends on DEBUG_KERNEL 1041 help 1042 If you say Y here, some extra kobject debugging messages will be sent 1043 to the syslog. 1044 1045config DEBUG_KOBJECT_RELEASE 1046 bool "kobject release debugging" 1047 depends on DEBUG_OBJECTS_TIMERS 1048 help 1049 kobjects are reference counted objects. This means that their 1050 last reference count put is not predictable, and the kobject can 1051 live on past the point at which a driver decides to drop it's 1052 initial reference to the kobject gained on allocation. An 1053 example of this would be a struct device which has just been 1054 unregistered. 1055 1056 However, some buggy drivers assume that after such an operation, 1057 the memory backing the kobject can be immediately freed. This 1058 goes completely against the principles of a refcounted object. 1059 1060 If you say Y here, the kernel will delay the release of kobjects 1061 on the last reference count to improve the visibility of this 1062 kind of kobject release bug. 1063 1064config HAVE_DEBUG_BUGVERBOSE 1065 bool 1066 1067config DEBUG_BUGVERBOSE 1068 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 1069 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE) 1070 default y 1071 help 1072 Say Y here to make BUG() panics output the file name and line number 1073 of the BUG call as well as the EIP and oops trace. This aids 1074 debugging but costs about 70-100K of memory. 1075 1076config DEBUG_LIST 1077 bool "Debug linked list manipulation" 1078 depends on DEBUG_KERNEL 1079 help 1080 Enable this to turn on extended checks in the linked-list 1081 walking routines. 1082 1083 If unsure, say N. 1084 1085config DEBUG_PI_LIST 1086 bool "Debug priority linked list manipulation" 1087 depends on DEBUG_KERNEL 1088 help 1089 Enable this to turn on extended checks in the priority-ordered 1090 linked-list (plist) walking routines. This checks the entire 1091 list multiple times during each manipulation. 1092 1093 If unsure, say N. 1094 1095config DEBUG_SG 1096 bool "Debug SG table operations" 1097 depends on DEBUG_KERNEL 1098 help 1099 Enable this to turn on checks on scatter-gather tables. This can 1100 help find problems with drivers that do not properly initialize 1101 their sg tables. 1102 1103 If unsure, say N. 1104 1105config DEBUG_NOTIFIERS 1106 bool "Debug notifier call chains" 1107 depends on DEBUG_KERNEL 1108 help 1109 Enable this to turn on sanity checking for notifier call chains. 1110 This is most useful for kernel developers to make sure that 1111 modules properly unregister themselves from notifier chains. 1112 This is a relatively cheap check but if you care about maximum 1113 performance, say N. 1114 1115config DEBUG_CREDENTIALS 1116 bool "Debug credential management" 1117 depends on DEBUG_KERNEL 1118 help 1119 Enable this to turn on some debug checking for credential 1120 management. The additional code keeps track of the number of 1121 pointers from task_structs to any given cred struct, and checks to 1122 see that this number never exceeds the usage count of the cred 1123 struct. 1124 1125 Furthermore, if SELinux is enabled, this also checks that the 1126 security pointer in the cred struct is never seen to be invalid. 1127 1128 If unsure, say N. 1129 1130menu "RCU Debugging" 1131 1132config PROVE_RCU 1133 bool "RCU debugging: prove RCU correctness" 1134 depends on PROVE_LOCKING 1135 default n 1136 help 1137 This feature enables lockdep extensions that check for correct 1138 use of RCU APIs. This is currently under development. Say Y 1139 if you want to debug RCU usage or help work on the PROVE_RCU 1140 feature. 1141 1142 Say N if you are unsure. 1143 1144config PROVE_RCU_REPEATEDLY 1145 bool "RCU debugging: don't disable PROVE_RCU on first splat" 1146 depends on PROVE_RCU 1147 default n 1148 help 1149 By itself, PROVE_RCU will disable checking upon issuing the 1150 first warning (or "splat"). This feature prevents such 1151 disabling, allowing multiple RCU-lockdep warnings to be printed 1152 on a single reboot. 1153 1154 Say Y to allow multiple RCU-lockdep warnings per boot. 1155 1156 Say N if you are unsure. 1157 1158config SPARSE_RCU_POINTER 1159 bool "RCU debugging: sparse-based checks for pointer usage" 1160 default n 1161 help 1162 This feature enables the __rcu sparse annotation for 1163 RCU-protected pointers. This annotation will cause sparse 1164 to flag any non-RCU used of annotated pointers. This can be 1165 helpful when debugging RCU usage. Please note that this feature 1166 is not intended to enforce code cleanliness; it is instead merely 1167 a debugging aid. 1168 1169 Say Y to make sparse flag questionable use of RCU-protected pointers 1170 1171 Say N if you are unsure. 1172 1173config TORTURE_TEST 1174 tristate 1175 default n 1176 1177config RCU_TORTURE_TEST 1178 tristate "torture tests for RCU" 1179 depends on DEBUG_KERNEL 1180 select TORTURE_TEST 1181 default n 1182 help 1183 This option provides a kernel module that runs torture tests 1184 on the RCU infrastructure. The kernel module may be built 1185 after the fact on the running kernel to be tested, if desired. 1186 1187 Say Y here if you want RCU torture tests to be built into 1188 the kernel. 1189 Say M if you want the RCU torture tests to build as a module. 1190 Say N if you are unsure. 1191 1192config RCU_TORTURE_TEST_RUNNABLE 1193 bool "torture tests for RCU runnable by default" 1194 depends on RCU_TORTURE_TEST = y 1195 default n 1196 help 1197 This option provides a way to build the RCU torture tests 1198 directly into the kernel without them starting up at boot 1199 time. You can use /proc/sys/kernel/rcutorture_runnable 1200 to manually override this setting. This /proc file is 1201 available only when the RCU torture tests have been built 1202 into the kernel. 1203 1204 Say Y here if you want the RCU torture tests to start during 1205 boot (you probably don't). 1206 Say N here if you want the RCU torture tests to start only 1207 after being manually enabled via /proc. 1208 1209config RCU_CPU_STALL_TIMEOUT 1210 int "RCU CPU stall timeout in seconds" 1211 depends on RCU_STALL_COMMON 1212 range 3 300 1213 default 21 1214 help 1215 If a given RCU grace period extends more than the specified 1216 number of seconds, a CPU stall warning is printed. If the 1217 RCU grace period persists, additional CPU stall warnings are 1218 printed at more widely spaced intervals. 1219 1220config RCU_CPU_STALL_VERBOSE 1221 bool "Print additional per-task information for RCU_CPU_STALL_DETECTOR" 1222 depends on TREE_PREEMPT_RCU 1223 default y 1224 help 1225 This option causes RCU to printk detailed per-task information 1226 for any tasks that are stalling the current RCU grace period. 1227 1228 Say N if you are unsure. 1229 1230 Say Y if you want to enable such checks. 1231 1232config RCU_CPU_STALL_INFO 1233 bool "Print additional diagnostics on RCU CPU stall" 1234 depends on (TREE_RCU || TREE_PREEMPT_RCU) && DEBUG_KERNEL 1235 default n 1236 help 1237 For each stalled CPU that is aware of the current RCU grace 1238 period, print out additional per-CPU diagnostic information 1239 regarding scheduling-clock ticks, idle state, and, 1240 for RCU_FAST_NO_HZ kernels, idle-entry state. 1241 1242 Say N if you are unsure. 1243 1244 Say Y if you want to enable such diagnostics. 1245 1246config RCU_TRACE 1247 bool "Enable tracing for RCU" 1248 depends on DEBUG_KERNEL 1249 select TRACE_CLOCK 1250 help 1251 This option provides tracing in RCU which presents stats 1252 in debugfs for debugging RCU implementation. 1253 1254 Say Y here if you want to enable RCU tracing 1255 Say N if you are unsure. 1256 1257endmenu # "RCU Debugging" 1258 1259config DEBUG_BLOCK_EXT_DEVT 1260 bool "Force extended block device numbers and spread them" 1261 depends on DEBUG_KERNEL 1262 depends on BLOCK 1263 default n 1264 help 1265 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1266 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1267 YOU ARE DOING. Distros, please enable this and fix whatever 1268 is broken. 1269 1270 Conventionally, block device numbers are allocated from 1271 predetermined contiguous area. However, extended block area 1272 may introduce non-contiguous block device numbers. This 1273 option forces most block device numbers to be allocated from 1274 the extended space and spreads them to discover kernel or 1275 userland code paths which assume predetermined contiguous 1276 device number allocation. 1277 1278 Note that turning on this debug option shuffles all the 1279 device numbers for all IDE and SCSI devices including libata 1280 ones, so root partition specified using device number 1281 directly (via rdev or root=MAJ:MIN) won't work anymore. 1282 Textual device names (root=/dev/sdXn) will continue to work. 1283 1284 Say N if you are unsure. 1285 1286config NOTIFIER_ERROR_INJECTION 1287 tristate "Notifier error injection" 1288 depends on DEBUG_KERNEL 1289 select DEBUG_FS 1290 help 1291 This option provides the ability to inject artificial errors to 1292 specified notifier chain callbacks. It is useful to test the error 1293 handling of notifier call chain failures. 1294 1295 Say N if unsure. 1296 1297config CPU_NOTIFIER_ERROR_INJECT 1298 tristate "CPU notifier error injection module" 1299 depends on HOTPLUG_CPU && NOTIFIER_ERROR_INJECTION 1300 help 1301 This option provides a kernel module that can be used to test 1302 the error handling of the cpu notifiers by injecting artificial 1303 errors to CPU notifier chain callbacks. It is controlled through 1304 debugfs interface under /sys/kernel/debug/notifier-error-inject/cpu 1305 1306 If the notifier call chain should be failed with some events 1307 notified, write the error code to "actions/<notifier event>/error". 1308 1309 Example: Inject CPU offline error (-1 == -EPERM) 1310 1311 # cd /sys/kernel/debug/notifier-error-inject/cpu 1312 # echo -1 > actions/CPU_DOWN_PREPARE/error 1313 # echo 0 > /sys/devices/system/cpu/cpu1/online 1314 bash: echo: write error: Operation not permitted 1315 1316 To compile this code as a module, choose M here: the module will 1317 be called cpu-notifier-error-inject. 1318 1319 If unsure, say N. 1320 1321config PM_NOTIFIER_ERROR_INJECT 1322 tristate "PM notifier error injection module" 1323 depends on PM && NOTIFIER_ERROR_INJECTION 1324 default m if PM_DEBUG 1325 help 1326 This option provides the ability to inject artificial errors to 1327 PM notifier chain callbacks. It is controlled through debugfs 1328 interface /sys/kernel/debug/notifier-error-inject/pm 1329 1330 If the notifier call chain should be failed with some events 1331 notified, write the error code to "actions/<notifier event>/error". 1332 1333 Example: Inject PM suspend error (-12 = -ENOMEM) 1334 1335 # cd /sys/kernel/debug/notifier-error-inject/pm/ 1336 # echo -12 > actions/PM_SUSPEND_PREPARE/error 1337 # echo mem > /sys/power/state 1338 bash: echo: write error: Cannot allocate memory 1339 1340 To compile this code as a module, choose M here: the module will 1341 be called pm-notifier-error-inject. 1342 1343 If unsure, say N. 1344 1345config OF_RECONFIG_NOTIFIER_ERROR_INJECT 1346 tristate "OF reconfig notifier error injection module" 1347 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 1348 help 1349 This option provides the ability to inject artificial errors to 1350 OF reconfig notifier chain callbacks. It is controlled 1351 through debugfs interface under 1352 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 1353 1354 If the notifier call chain should be failed with some events 1355 notified, write the error code to "actions/<notifier event>/error". 1356 1357 To compile this code as a module, choose M here: the module will 1358 be called of-reconfig-notifier-error-inject. 1359 1360 If unsure, say N. 1361 1362config FAULT_INJECTION 1363 bool "Fault-injection framework" 1364 depends on DEBUG_KERNEL 1365 help 1366 Provide fault-injection framework. 1367 For more details, see Documentation/fault-injection/. 1368 1369config FAILSLAB 1370 bool "Fault-injection capability for kmalloc" 1371 depends on FAULT_INJECTION 1372 depends on SLAB || SLUB 1373 help 1374 Provide fault-injection capability for kmalloc. 1375 1376config FAIL_PAGE_ALLOC 1377 bool "Fault-injection capabilitiy for alloc_pages()" 1378 depends on FAULT_INJECTION 1379 help 1380 Provide fault-injection capability for alloc_pages(). 1381 1382config FAIL_MAKE_REQUEST 1383 bool "Fault-injection capability for disk IO" 1384 depends on FAULT_INJECTION && BLOCK 1385 help 1386 Provide fault-injection capability for disk IO. 1387 1388config FAIL_IO_TIMEOUT 1389 bool "Fault-injection capability for faking disk interrupts" 1390 depends on FAULT_INJECTION && BLOCK 1391 help 1392 Provide fault-injection capability on end IO handling. This 1393 will make the block layer "forget" an interrupt as configured, 1394 thus exercising the error handling. 1395 1396 Only works with drivers that use the generic timeout handling, 1397 for others it wont do anything. 1398 1399config FAIL_MMC_REQUEST 1400 bool "Fault-injection capability for MMC IO" 1401 select DEBUG_FS 1402 depends on FAULT_INJECTION && MMC 1403 help 1404 Provide fault-injection capability for MMC IO. 1405 This will make the mmc core return data errors. This is 1406 useful to test the error handling in the mmc block device 1407 and to test how the mmc host driver handles retries from 1408 the block device. 1409 1410config FAULT_INJECTION_DEBUG_FS 1411 bool "Debugfs entries for fault-injection capabilities" 1412 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1413 help 1414 Enable configuration of fault-injection capabilities via debugfs. 1415 1416config FAULT_INJECTION_STACKTRACE_FILTER 1417 bool "stacktrace filter for fault-injection capabilities" 1418 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1419 depends on !X86_64 1420 select STACKTRACE 1421 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !SCORE 1422 help 1423 Provide stacktrace filter for fault-injection capabilities 1424 1425config LATENCYTOP 1426 bool "Latency measuring infrastructure" 1427 depends on HAVE_LATENCYTOP_SUPPORT 1428 depends on DEBUG_KERNEL 1429 depends on STACKTRACE_SUPPORT 1430 depends on PROC_FS 1431 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC 1432 select KALLSYMS 1433 select KALLSYMS_ALL 1434 select STACKTRACE 1435 select SCHEDSTATS 1436 select SCHED_DEBUG 1437 help 1438 Enable this option if you want to use the LatencyTOP tool 1439 to find out which userspace is blocking on what kernel operations. 1440 1441config ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS 1442 bool 1443 1444config DEBUG_STRICT_USER_COPY_CHECKS 1445 bool "Strict user copy size checks" 1446 depends on ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS 1447 depends on DEBUG_KERNEL && !TRACE_BRANCH_PROFILING 1448 help 1449 Enabling this option turns a certain set of sanity checks for user 1450 copy operations into compile time failures. 1451 1452 The copy_from_user() etc checks are there to help test if there 1453 are sufficient security checks on the length argument of 1454 the copy operation, by having gcc prove that the argument is 1455 within bounds. 1456 1457 If unsure, say N. 1458 1459source kernel/trace/Kconfig 1460 1461menu "Runtime Testing" 1462 1463config LKDTM 1464 tristate "Linux Kernel Dump Test Tool Module" 1465 depends on DEBUG_FS 1466 depends on BLOCK 1467 default n 1468 help 1469 This module enables testing of the different dumping mechanisms by 1470 inducing system failures at predefined crash points. 1471 If you don't need it: say N 1472 Choose M here to compile this code as a module. The module will be 1473 called lkdtm. 1474 1475 Documentation on how to use the module can be found in 1476 Documentation/fault-injection/provoke-crashes.txt 1477 1478config TEST_LIST_SORT 1479 bool "Linked list sorting test" 1480 depends on DEBUG_KERNEL 1481 help 1482 Enable this to turn on 'list_sort()' function test. This test is 1483 executed only once during system boot, so affects only boot time. 1484 1485 If unsure, say N. 1486 1487config KPROBES_SANITY_TEST 1488 bool "Kprobes sanity tests" 1489 depends on DEBUG_KERNEL 1490 depends on KPROBES 1491 default n 1492 help 1493 This option provides for testing basic kprobes functionality on 1494 boot. A sample kprobe, jprobe and kretprobe are inserted and 1495 verified for functionality. 1496 1497 Say N if you are unsure. 1498 1499config BACKTRACE_SELF_TEST 1500 tristate "Self test for the backtrace code" 1501 depends on DEBUG_KERNEL 1502 default n 1503 help 1504 This option provides a kernel module that can be used to test 1505 the kernel stack backtrace code. This option is not useful 1506 for distributions or general kernels, but only for kernel 1507 developers working on architecture code. 1508 1509 Note that if you want to also test saved backtraces, you will 1510 have to enable STACKTRACE as well. 1511 1512 Say N if you are unsure. 1513 1514config RBTREE_TEST 1515 tristate "Red-Black tree test" 1516 depends on DEBUG_KERNEL 1517 help 1518 A benchmark measuring the performance of the rbtree library. 1519 Also includes rbtree invariant checks. 1520 1521config INTERVAL_TREE_TEST 1522 tristate "Interval tree test" 1523 depends on m && DEBUG_KERNEL 1524 select INTERVAL_TREE 1525 help 1526 A benchmark measuring the performance of the interval tree library 1527 1528config PERCPU_TEST 1529 tristate "Per cpu operations test" 1530 depends on m && DEBUG_KERNEL 1531 help 1532 Enable this option to build test module which validates per-cpu 1533 operations. 1534 1535 If unsure, say N. 1536 1537config ATOMIC64_SELFTEST 1538 bool "Perform an atomic64_t self-test at boot" 1539 help 1540 Enable this option to test the atomic64_t functions at boot. 1541 1542 If unsure, say N. 1543 1544config ASYNC_RAID6_TEST 1545 tristate "Self test for hardware accelerated raid6 recovery" 1546 depends on ASYNC_RAID6_RECOV 1547 select ASYNC_MEMCPY 1548 ---help--- 1549 This is a one-shot self test that permutes through the 1550 recovery of all the possible two disk failure scenarios for a 1551 N-disk array. Recovery is performed with the asynchronous 1552 raid6 recovery routines, and will optionally use an offload 1553 engine if one is available. 1554 1555 If unsure, say N. 1556 1557config TEST_STRING_HELPERS 1558 tristate "Test functions located in the string_helpers module at runtime" 1559 1560config TEST_KSTRTOX 1561 tristate "Test kstrto*() family of functions at runtime" 1562 1563config TEST_RHASHTABLE 1564 bool "Perform selftest on resizable hash table" 1565 default n 1566 help 1567 Enable this option to test the rhashtable functions at boot. 1568 1569 If unsure, say N. 1570 1571endmenu # runtime tests 1572 1573config PROVIDE_OHCI1394_DMA_INIT 1574 bool "Remote debugging over FireWire early on boot" 1575 depends on PCI && X86 1576 help 1577 If you want to debug problems which hang or crash the kernel early 1578 on boot and the crashing machine has a FireWire port, you can use 1579 this feature to remotely access the memory of the crashed machine 1580 over FireWire. This employs remote DMA as part of the OHCI1394 1581 specification which is now the standard for FireWire controllers. 1582 1583 With remote DMA, you can monitor the printk buffer remotely using 1584 firescope and access all memory below 4GB using fireproxy from gdb. 1585 Even controlling a kernel debugger is possible using remote DMA. 1586 1587 Usage: 1588 1589 If ohci1394_dma=early is used as boot parameter, it will initialize 1590 all OHCI1394 controllers which are found in the PCI config space. 1591 1592 As all changes to the FireWire bus such as enabling and disabling 1593 devices cause a bus reset and thereby disable remote DMA for all 1594 devices, be sure to have the cable plugged and FireWire enabled on 1595 the debugging host before booting the debug target for debugging. 1596 1597 This code (~1k) is freed after boot. By then, the firewire stack 1598 in charge of the OHCI-1394 controllers should be used instead. 1599 1600 See Documentation/debugging-via-ohci1394.txt for more information. 1601 1602config BUILD_DOCSRC 1603 bool "Build targets in Documentation/ tree" 1604 depends on HEADERS_CHECK 1605 help 1606 This option attempts to build objects from the source files in the 1607 kernel Documentation/ tree. 1608 1609 Say N if you are unsure. 1610 1611config DMA_API_DEBUG 1612 bool "Enable debugging of DMA-API usage" 1613 depends on HAVE_DMA_API_DEBUG 1614 help 1615 Enable this option to debug the use of the DMA API by device drivers. 1616 With this option you will be able to detect common bugs in device 1617 drivers like double-freeing of DMA mappings or freeing mappings that 1618 were never allocated. 1619 1620 This also attempts to catch cases where a page owned by DMA is 1621 accessed by the cpu in a way that could cause data corruption. For 1622 example, this enables cow_user_page() to check that the source page is 1623 not undergoing DMA. 1624 1625 This option causes a performance degradation. Use only if you want to 1626 debug device drivers and dma interactions. 1627 1628 If unsure, say N. 1629 1630config TEST_MODULE 1631 tristate "Test module loading with 'hello world' module" 1632 default n 1633 depends on m 1634 help 1635 This builds the "test_module" module that emits "Hello, world" 1636 on printk when loaded. It is designed to be used for basic 1637 evaluation of the module loading subsystem (for example when 1638 validating module verification). It lacks any extra dependencies, 1639 and will not normally be loaded by the system unless explicitly 1640 requested by name. 1641 1642 If unsure, say N. 1643 1644config TEST_USER_COPY 1645 tristate "Test user/kernel boundary protections" 1646 default n 1647 depends on m 1648 help 1649 This builds the "test_user_copy" module that runs sanity checks 1650 on the copy_to/from_user infrastructure, making sure basic 1651 user/kernel boundary testing is working. If it fails to load, 1652 a regression has been detected in the user/kernel memory boundary 1653 protections. 1654 1655 If unsure, say N. 1656 1657config TEST_BPF 1658 tristate "Test BPF filter functionality" 1659 default n 1660 depends on m && NET 1661 help 1662 This builds the "test_bpf" module that runs various test vectors 1663 against the BPF interpreter or BPF JIT compiler depending on the 1664 current setting. This is in particular useful for BPF JIT compiler 1665 development, but also to run regression tests against changes in 1666 the interpreter code. 1667 1668 If unsure, say N. 1669 1670config TEST_FIRMWARE 1671 tristate "Test firmware loading via userspace interface" 1672 default n 1673 depends on FW_LOADER 1674 help 1675 This builds the "test_firmware" module that creates a userspace 1676 interface for testing firmware loading. This can be used to 1677 control the triggering of firmware loading without needing an 1678 actual firmware-using device. The contents can be rechecked by 1679 userspace. 1680 1681 If unsure, say N. 1682 1683config TEST_UDELAY 1684 tristate "udelay test driver" 1685 default n 1686 help 1687 This builds the "udelay_test" module that helps to make sure 1688 that udelay() is working properly. 1689 1690 If unsure, say N. 1691 1692source "samples/Kconfig" 1693 1694source "lib/Kconfig.kgdb" 1695 1696