1 2config PRINTK_TIME 3 bool "Show timing information on printks" 4 depends on PRINTK 5 help 6 Selecting this option causes timing information to be 7 included in printk output. This allows you to measure 8 the interval between kernel operations, including bootup 9 operations. This is useful for identifying long delays 10 in kernel startup. Or add printk.time=1 at boot-time. 11 See Documentation/kernel-parameters.txt 12 13config DEFAULT_MESSAGE_LOGLEVEL 14 int "Default message log level (1-7)" 15 range 1 7 16 default "4" 17 help 18 Default log level for printk statements with no specified priority. 19 20 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks 21 that are auditing their logs closely may want to set it to a lower 22 priority. 23 24config ENABLE_WARN_DEPRECATED 25 bool "Enable __deprecated logic" 26 default y 27 help 28 Enable the __deprecated logic in the kernel build. 29 Disable this to suppress the "warning: 'foo' is deprecated 30 (declared at kernel/power/somefile.c:1234)" messages. 31 32config ENABLE_MUST_CHECK 33 bool "Enable __must_check logic" 34 default y 35 help 36 Enable the __must_check logic in the kernel build. Disable this to 37 suppress the "warning: ignoring return value of 'foo', declared with 38 attribute warn_unused_result" messages. 39 40config FRAME_WARN 41 int "Warn for stack frames larger than (needs gcc 4.4)" 42 range 0 8192 43 default 1024 if !64BIT 44 default 2048 if 64BIT 45 help 46 Tell gcc to warn at build time for stack frames larger than this. 47 Setting this too low will cause a lot of warnings. 48 Setting it to 0 disables the warning. 49 Requires gcc 4.4 50 51config MAGIC_SYSRQ 52 bool "Magic SysRq key" 53 depends on !UML 54 help 55 If you say Y here, you will have some control over the system even 56 if the system crashes for example during kernel debugging (e.g., you 57 will be able to flush the buffer cache to disk, reboot the system 58 immediately or dump some status information). This is accomplished 59 by pressing various keys while holding SysRq (Alt+PrintScreen). It 60 also works on a serial console (on PC hardware at least), if you 61 send a BREAK and then within 5 seconds a command keypress. The 62 keys are documented in <file:Documentation/sysrq.txt>. Don't say Y 63 unless you really know what this hack does. 64 65config STRIP_ASM_SYMS 66 bool "Strip assembler-generated symbols during link" 67 default n 68 help 69 Strip internal assembler-generated symbols during a link (symbols 70 that look like '.Lxxx') so they don't pollute the output of 71 get_wchan() and suchlike. 72 73config UNUSED_SYMBOLS 74 bool "Enable unused/obsolete exported symbols" 75 default y if X86 76 help 77 Unused but exported symbols make the kernel needlessly bigger. For 78 that reason most of these unused exports will soon be removed. This 79 option is provided temporarily to provide a transition period in case 80 some external kernel module needs one of these symbols anyway. If you 81 encounter such a case in your module, consider if you are actually 82 using the right API. (rationale: since nobody in the kernel is using 83 this in a module, there is a pretty good chance it's actually the 84 wrong interface to use). If you really need the symbol, please send a 85 mail to the linux kernel mailing list mentioning the symbol and why 86 you really need it, and what the merge plan to the mainline kernel for 87 your module is. 88 89config DEBUG_FS 90 bool "Debug Filesystem" 91 help 92 debugfs is a virtual file system that kernel developers use to put 93 debugging files into. Enable this option to be able to read and 94 write to these files. 95 96 For detailed documentation on the debugfs API, see 97 Documentation/DocBook/filesystems. 98 99 If unsure, say N. 100 101config HEADERS_CHECK 102 bool "Run 'make headers_check' when building vmlinux" 103 depends on !UML 104 help 105 This option will extract the user-visible kernel headers whenever 106 building the kernel, and will run basic sanity checks on them to 107 ensure that exported files do not attempt to include files which 108 were not exported, etc. 109 110 If you're making modifications to header files which are 111 relevant for userspace, say 'Y', and check the headers 112 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in 113 your build tree), to make sure they're suitable. 114 115config DEBUG_SECTION_MISMATCH 116 bool "Enable full Section mismatch analysis" 117 help 118 The section mismatch analysis checks if there are illegal 119 references from one section to another section. 120 During linktime or runtime, some sections are dropped; 121 any use of code/data previously in these sections would 122 most likely result in an oops. 123 In the code, functions and variables are annotated with 124 __init, __devinit, etc. (see the full list in include/linux/init.h), 125 which results in the code/data being placed in specific sections. 126 The section mismatch analysis is always performed after a full 127 kernel build, and enabling this option causes the following 128 additional steps to occur: 129 - Add the option -fno-inline-functions-called-once to gcc commands. 130 When inlining a function annotated with __init in a non-init 131 function, we would lose the section information and thus 132 the analysis would not catch the illegal reference. 133 This option tells gcc to inline less (but it does result in 134 a larger kernel). 135 - Run the section mismatch analysis for each module/built-in.o file. 136 When we run the section mismatch analysis on vmlinux.o, we 137 lose valueble information about where the mismatch was 138 introduced. 139 Running the analysis for each module/built-in.o file 140 tells where the mismatch happens much closer to the 141 source. The drawback is that the same mismatch is 142 reported at least twice. 143 - Enable verbose reporting from modpost in order to help resolve 144 the section mismatches that are reported. 145 146config DEBUG_KERNEL 147 bool "Kernel debugging" 148 help 149 Say Y here if you are developing drivers or trying to debug and 150 identify kernel problems. 151 152config DEBUG_SHIRQ 153 bool "Debug shared IRQ handlers" 154 depends on DEBUG_KERNEL && GENERIC_HARDIRQS 155 help 156 Enable this to generate a spurious interrupt as soon as a shared 157 interrupt handler is registered, and just before one is deregistered. 158 Drivers ought to be able to handle interrupts coming in at those 159 points; some don't and need to be caught. 160 161config LOCKUP_DETECTOR 162 bool "Detect Hard and Soft Lockups" 163 depends on DEBUG_KERNEL && !S390 164 help 165 Say Y here to enable the kernel to act as a watchdog to detect 166 hard and soft lockups. 167 168 Softlockups are bugs that cause the kernel to loop in kernel 169 mode for more than 60 seconds, without giving other tasks a 170 chance to run. The current stack trace is displayed upon 171 detection and the system will stay locked up. 172 173 Hardlockups are bugs that cause the CPU to loop in kernel mode 174 for more than 60 seconds, without letting other interrupts have a 175 chance to run. The current stack trace is displayed upon detection 176 and the system will stay locked up. 177 178 The overhead should be minimal. A periodic hrtimer runs to 179 generate interrupts and kick the watchdog task every 10-12 seconds. 180 An NMI is generated every 60 seconds or so to check for hardlockups. 181 182config HARDLOCKUP_DETECTOR 183 def_bool LOCKUP_DETECTOR && PERF_EVENTS && HAVE_PERF_EVENTS_NMI && \ 184 !ARCH_HAS_NMI_WATCHDOG 185 186config BOOTPARAM_HARDLOCKUP_PANIC 187 bool "Panic (Reboot) On Hard Lockups" 188 depends on LOCKUP_DETECTOR 189 help 190 Say Y here to enable the kernel to panic on "hard lockups", 191 which are bugs that cause the kernel to loop in kernel 192 mode with interrupts disabled for more than 60 seconds. 193 194 Say N if unsure. 195 196config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 197 int 198 depends on LOCKUP_DETECTOR 199 range 0 1 200 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 201 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 202 203config BOOTPARAM_SOFTLOCKUP_PANIC 204 bool "Panic (Reboot) On Soft Lockups" 205 depends on LOCKUP_DETECTOR 206 help 207 Say Y here to enable the kernel to panic on "soft lockups", 208 which are bugs that cause the kernel to loop in kernel 209 mode for more than 60 seconds, without giving other tasks a 210 chance to run. 211 212 The panic can be used in combination with panic_timeout, 213 to cause the system to reboot automatically after a 214 lockup has been detected. This feature is useful for 215 high-availability systems that have uptime guarantees and 216 where a lockup must be resolved ASAP. 217 218 Say N if unsure. 219 220config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 221 int 222 depends on LOCKUP_DETECTOR 223 range 0 1 224 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 225 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 226 227config DETECT_HUNG_TASK 228 bool "Detect Hung Tasks" 229 depends on DEBUG_KERNEL 230 default LOCKUP_DETECTOR 231 help 232 Say Y here to enable the kernel to detect "hung tasks", 233 which are bugs that cause the task to be stuck in 234 uninterruptible "D" state indefinitiley. 235 236 When a hung task is detected, the kernel will print the 237 current stack trace (which you should report), but the 238 task will stay in uninterruptible state. If lockdep is 239 enabled then all held locks will also be reported. This 240 feature has negligible overhead. 241 242config DEFAULT_HUNG_TASK_TIMEOUT 243 int "Default timeout for hung task detection (in seconds)" 244 depends on DETECT_HUNG_TASK 245 default 120 246 help 247 This option controls the default timeout (in seconds) used 248 to determine when a task has become non-responsive and should 249 be considered hung. 250 251 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 252 sysctl or by writing a value to 253 /proc/sys/kernel/hung_task_timeout_secs. 254 255 A timeout of 0 disables the check. The default is two minutes. 256 Keeping the default should be fine in most cases. 257 258config BOOTPARAM_HUNG_TASK_PANIC 259 bool "Panic (Reboot) On Hung Tasks" 260 depends on DETECT_HUNG_TASK 261 help 262 Say Y here to enable the kernel to panic on "hung tasks", 263 which are bugs that cause the kernel to leave a task stuck 264 in uninterruptible "D" state. 265 266 The panic can be used in combination with panic_timeout, 267 to cause the system to reboot automatically after a 268 hung task has been detected. This feature is useful for 269 high-availability systems that have uptime guarantees and 270 where a hung tasks must be resolved ASAP. 271 272 Say N if unsure. 273 274config BOOTPARAM_HUNG_TASK_PANIC_VALUE 275 int 276 depends on DETECT_HUNG_TASK 277 range 0 1 278 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 279 default 1 if BOOTPARAM_HUNG_TASK_PANIC 280 281config SCHED_DEBUG 282 bool "Collect scheduler debugging info" 283 depends on DEBUG_KERNEL && PROC_FS 284 default y 285 help 286 If you say Y here, the /proc/sched_debug file will be provided 287 that can help debug the scheduler. The runtime overhead of this 288 option is minimal. 289 290config SCHEDSTATS 291 bool "Collect scheduler statistics" 292 depends on DEBUG_KERNEL && PROC_FS 293 help 294 If you say Y here, additional code will be inserted into the 295 scheduler and related routines to collect statistics about 296 scheduler behavior and provide them in /proc/schedstat. These 297 stats may be useful for both tuning and debugging the scheduler 298 If you aren't debugging the scheduler or trying to tune a specific 299 application, you can say N to avoid the very slight overhead 300 this adds. 301 302config TIMER_STATS 303 bool "Collect kernel timers statistics" 304 depends on DEBUG_KERNEL && PROC_FS 305 help 306 If you say Y here, additional code will be inserted into the 307 timer routines to collect statistics about kernel timers being 308 reprogrammed. The statistics can be read from /proc/timer_stats. 309 The statistics collection is started by writing 1 to /proc/timer_stats, 310 writing 0 stops it. This feature is useful to collect information 311 about timer usage patterns in kernel and userspace. This feature 312 is lightweight if enabled in the kernel config but not activated 313 (it defaults to deactivated on bootup and will only be activated 314 if some application like powertop activates it explicitly). 315 316config DEBUG_OBJECTS 317 bool "Debug object operations" 318 depends on DEBUG_KERNEL 319 help 320 If you say Y here, additional code will be inserted into the 321 kernel to track the life time of various objects and validate 322 the operations on those objects. 323 324config DEBUG_OBJECTS_SELFTEST 325 bool "Debug objects selftest" 326 depends on DEBUG_OBJECTS 327 help 328 This enables the selftest of the object debug code. 329 330config DEBUG_OBJECTS_FREE 331 bool "Debug objects in freed memory" 332 depends on DEBUG_OBJECTS 333 help 334 This enables checks whether a k/v free operation frees an area 335 which contains an object which has not been deactivated 336 properly. This can make kmalloc/kfree-intensive workloads 337 much slower. 338 339config DEBUG_OBJECTS_TIMERS 340 bool "Debug timer objects" 341 depends on DEBUG_OBJECTS 342 help 343 If you say Y here, additional code will be inserted into the 344 timer routines to track the life time of timer objects and 345 validate the timer operations. 346 347config DEBUG_OBJECTS_WORK 348 bool "Debug work objects" 349 depends on DEBUG_OBJECTS 350 help 351 If you say Y here, additional code will be inserted into the 352 work queue routines to track the life time of work objects and 353 validate the work operations. 354 355config DEBUG_OBJECTS_RCU_HEAD 356 bool "Debug RCU callbacks objects" 357 depends on DEBUG_OBJECTS 358 help 359 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 360 361config DEBUG_OBJECTS_PERCPU_COUNTER 362 bool "Debug percpu counter objects" 363 depends on DEBUG_OBJECTS 364 help 365 If you say Y here, additional code will be inserted into the 366 percpu counter routines to track the life time of percpu counter 367 objects and validate the percpu counter operations. 368 369config DEBUG_OBJECTS_ENABLE_DEFAULT 370 int "debug_objects bootup default value (0-1)" 371 range 0 1 372 default "1" 373 depends on DEBUG_OBJECTS 374 help 375 Debug objects boot parameter default value 376 377config DEBUG_SLAB 378 bool "Debug slab memory allocations" 379 depends on DEBUG_KERNEL && SLAB && !KMEMCHECK 380 help 381 Say Y here to have the kernel do limited verification on memory 382 allocation as well as poisoning memory on free to catch use of freed 383 memory. This can make kmalloc/kfree-intensive workloads much slower. 384 385config DEBUG_SLAB_LEAK 386 bool "Memory leak debugging" 387 depends on DEBUG_SLAB 388 389config SLUB_DEBUG_ON 390 bool "SLUB debugging on by default" 391 depends on SLUB && SLUB_DEBUG && !KMEMCHECK 392 default n 393 help 394 Boot with debugging on by default. SLUB boots by default with 395 the runtime debug capabilities switched off. Enabling this is 396 equivalent to specifying the "slub_debug" parameter on boot. 397 There is no support for more fine grained debug control like 398 possible with slub_debug=xxx. SLUB debugging may be switched 399 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 400 "slub_debug=-". 401 402config SLUB_STATS 403 default n 404 bool "Enable SLUB performance statistics" 405 depends on SLUB && SYSFS 406 help 407 SLUB statistics are useful to debug SLUBs allocation behavior in 408 order find ways to optimize the allocator. This should never be 409 enabled for production use since keeping statistics slows down 410 the allocator by a few percentage points. The slabinfo command 411 supports the determination of the most active slabs to figure 412 out which slabs are relevant to a particular load. 413 Try running: slabinfo -DA 414 415config DEBUG_KMEMLEAK 416 bool "Kernel memory leak detector" 417 depends on DEBUG_KERNEL && EXPERIMENTAL && \ 418 (X86 || ARM || PPC || MIPS || S390 || SPARC64 || SUPERH || MICROBLAZE || TILE) 419 420 select DEBUG_FS 421 select STACKTRACE if STACKTRACE_SUPPORT 422 select KALLSYMS 423 select CRC32 424 help 425 Say Y here if you want to enable the memory leak 426 detector. The memory allocation/freeing is traced in a way 427 similar to the Boehm's conservative garbage collector, the 428 difference being that the orphan objects are not freed but 429 only shown in /sys/kernel/debug/kmemleak. Enabling this 430 feature will introduce an overhead to memory 431 allocations. See Documentation/kmemleak.txt for more 432 details. 433 434 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 435 of finding leaks due to the slab objects poisoning. 436 437 In order to access the kmemleak file, debugfs needs to be 438 mounted (usually at /sys/kernel/debug). 439 440config DEBUG_KMEMLEAK_EARLY_LOG_SIZE 441 int "Maximum kmemleak early log entries" 442 depends on DEBUG_KMEMLEAK 443 range 200 40000 444 default 400 445 help 446 Kmemleak must track all the memory allocations to avoid 447 reporting false positives. Since memory may be allocated or 448 freed before kmemleak is initialised, an early log buffer is 449 used to store these actions. If kmemleak reports "early log 450 buffer exceeded", please increase this value. 451 452config DEBUG_KMEMLEAK_TEST 453 tristate "Simple test for the kernel memory leak detector" 454 depends on DEBUG_KMEMLEAK && m 455 help 456 This option enables a module that explicitly leaks memory. 457 458 If unsure, say N. 459 460config DEBUG_KMEMLEAK_DEFAULT_OFF 461 bool "Default kmemleak to off" 462 depends on DEBUG_KMEMLEAK 463 help 464 Say Y here to disable kmemleak by default. It can then be enabled 465 on the command line via kmemleak=on. 466 467config DEBUG_PREEMPT 468 bool "Debug preemptible kernel" 469 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT 470 default y 471 help 472 If you say Y here then the kernel will use a debug variant of the 473 commonly used smp_processor_id() function and will print warnings 474 if kernel code uses it in a preemption-unsafe way. Also, the kernel 475 will detect preemption count underflows. 476 477config DEBUG_RT_MUTEXES 478 bool "RT Mutex debugging, deadlock detection" 479 depends on DEBUG_KERNEL && RT_MUTEXES 480 help 481 This allows rt mutex semantics violations and rt mutex related 482 deadlocks (lockups) to be detected and reported automatically. 483 484config DEBUG_PI_LIST 485 bool 486 default y 487 depends on DEBUG_RT_MUTEXES 488 489config RT_MUTEX_TESTER 490 bool "Built-in scriptable tester for rt-mutexes" 491 depends on DEBUG_KERNEL && RT_MUTEXES 492 help 493 This option enables a rt-mutex tester. 494 495config DEBUG_SPINLOCK 496 bool "Spinlock and rw-lock debugging: basic checks" 497 depends on DEBUG_KERNEL 498 help 499 Say Y here and build SMP to catch missing spinlock initialization 500 and certain other kinds of spinlock errors commonly made. This is 501 best used in conjunction with the NMI watchdog so that spinlock 502 deadlocks are also debuggable. 503 504config DEBUG_MUTEXES 505 bool "Mutex debugging: basic checks" 506 depends on DEBUG_KERNEL 507 help 508 This feature allows mutex semantics violations to be detected and 509 reported. 510 511config DEBUG_LOCK_ALLOC 512 bool "Lock debugging: detect incorrect freeing of live locks" 513 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 514 select DEBUG_SPINLOCK 515 select DEBUG_MUTEXES 516 select LOCKDEP 517 help 518 This feature will check whether any held lock (spinlock, rwlock, 519 mutex or rwsem) is incorrectly freed by the kernel, via any of the 520 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 521 vfree(), etc.), whether a live lock is incorrectly reinitialized via 522 spin_lock_init()/mutex_init()/etc., or whether there is any lock 523 held during task exit. 524 525config PROVE_LOCKING 526 bool "Lock debugging: prove locking correctness" 527 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 528 select LOCKDEP 529 select DEBUG_SPINLOCK 530 select DEBUG_MUTEXES 531 select DEBUG_LOCK_ALLOC 532 select TRACE_IRQFLAGS 533 default n 534 help 535 This feature enables the kernel to prove that all locking 536 that occurs in the kernel runtime is mathematically 537 correct: that under no circumstance could an arbitrary (and 538 not yet triggered) combination of observed locking 539 sequences (on an arbitrary number of CPUs, running an 540 arbitrary number of tasks and interrupt contexts) cause a 541 deadlock. 542 543 In short, this feature enables the kernel to report locking 544 related deadlocks before they actually occur. 545 546 The proof does not depend on how hard and complex a 547 deadlock scenario would be to trigger: how many 548 participant CPUs, tasks and irq-contexts would be needed 549 for it to trigger. The proof also does not depend on 550 timing: if a race and a resulting deadlock is possible 551 theoretically (no matter how unlikely the race scenario 552 is), it will be proven so and will immediately be 553 reported by the kernel (once the event is observed that 554 makes the deadlock theoretically possible). 555 556 If a deadlock is impossible (i.e. the locking rules, as 557 observed by the kernel, are mathematically correct), the 558 kernel reports nothing. 559 560 NOTE: this feature can also be enabled for rwlocks, mutexes 561 and rwsems - in which case all dependencies between these 562 different locking variants are observed and mapped too, and 563 the proof of observed correctness is also maintained for an 564 arbitrary combination of these separate locking variants. 565 566 For more details, see Documentation/lockdep-design.txt. 567 568config PROVE_RCU 569 bool "RCU debugging: prove RCU correctness" 570 depends on PROVE_LOCKING 571 default n 572 help 573 This feature enables lockdep extensions that check for correct 574 use of RCU APIs. This is currently under development. Say Y 575 if you want to debug RCU usage or help work on the PROVE_RCU 576 feature. 577 578 Say N if you are unsure. 579 580config PROVE_RCU_REPEATEDLY 581 bool "RCU debugging: don't disable PROVE_RCU on first splat" 582 depends on PROVE_RCU 583 default n 584 help 585 By itself, PROVE_RCU will disable checking upon issuing the 586 first warning (or "splat"). This feature prevents such 587 disabling, allowing multiple RCU-lockdep warnings to be printed 588 on a single reboot. 589 590 Say Y to allow multiple RCU-lockdep warnings per boot. 591 592 Say N if you are unsure. 593 594config SPARSE_RCU_POINTER 595 bool "RCU debugging: sparse-based checks for pointer usage" 596 default n 597 help 598 This feature enables the __rcu sparse annotation for 599 RCU-protected pointers. This annotation will cause sparse 600 to flag any non-RCU used of annotated pointers. This can be 601 helpful when debugging RCU usage. Please note that this feature 602 is not intended to enforce code cleanliness; it is instead merely 603 a debugging aid. 604 605 Say Y to make sparse flag questionable use of RCU-protected pointers 606 607 Say N if you are unsure. 608 609config LOCKDEP 610 bool 611 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 612 select STACKTRACE 613 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE 614 select KALLSYMS 615 select KALLSYMS_ALL 616 617config LOCK_STAT 618 bool "Lock usage statistics" 619 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 620 select LOCKDEP 621 select DEBUG_SPINLOCK 622 select DEBUG_MUTEXES 623 select DEBUG_LOCK_ALLOC 624 default n 625 help 626 This feature enables tracking lock contention points 627 628 For more details, see Documentation/lockstat.txt 629 630 This also enables lock events required by "perf lock", 631 subcommand of perf. 632 If you want to use "perf lock", you also need to turn on 633 CONFIG_EVENT_TRACING. 634 635 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 636 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 637 638config DEBUG_LOCKDEP 639 bool "Lock dependency engine debugging" 640 depends on DEBUG_KERNEL && LOCKDEP 641 help 642 If you say Y here, the lock dependency engine will do 643 additional runtime checks to debug itself, at the price 644 of more runtime overhead. 645 646config TRACE_IRQFLAGS 647 bool 648 help 649 Enables hooks to interrupt enabling and disabling for 650 either tracing or lock debugging. 651 652config DEBUG_ATOMIC_SLEEP 653 bool "Sleep inside atomic section checking" 654 select PREEMPT_COUNT 655 depends on DEBUG_KERNEL 656 help 657 If you say Y here, various routines which may sleep will become very 658 noisy if they are called inside atomic sections: when a spinlock is 659 held, inside an rcu read side critical section, inside preempt disabled 660 sections, inside an interrupt, etc... 661 662config DEBUG_LOCKING_API_SELFTESTS 663 bool "Locking API boot-time self-tests" 664 depends on DEBUG_KERNEL 665 help 666 Say Y here if you want the kernel to run a short self-test during 667 bootup. The self-test checks whether common types of locking bugs 668 are detected by debugging mechanisms or not. (if you disable 669 lock debugging then those bugs wont be detected of course.) 670 The following locking APIs are covered: spinlocks, rwlocks, 671 mutexes and rwsems. 672 673config STACKTRACE 674 bool 675 depends on STACKTRACE_SUPPORT 676 677config DEBUG_STACK_USAGE 678 bool "Stack utilization instrumentation" 679 depends on DEBUG_KERNEL 680 help 681 Enables the display of the minimum amount of free stack which each 682 task has ever had available in the sysrq-T and sysrq-P debug output. 683 684 This option will slow down process creation somewhat. 685 686config DEBUG_KOBJECT 687 bool "kobject debugging" 688 depends on DEBUG_KERNEL 689 help 690 If you say Y here, some extra kobject debugging messages will be sent 691 to the syslog. 692 693config DEBUG_HIGHMEM 694 bool "Highmem debugging" 695 depends on DEBUG_KERNEL && HIGHMEM 696 help 697 This options enables addition error checking for high memory systems. 698 Disable for production systems. 699 700config DEBUG_BUGVERBOSE 701 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 702 depends on BUG 703 depends on ARM || AVR32 || M32R || M68K || SPARC32 || SPARC64 || \ 704 FRV || SUPERH || GENERIC_BUG || BLACKFIN || MN10300 || TILE 705 default y 706 help 707 Say Y here to make BUG() panics output the file name and line number 708 of the BUG call as well as the EIP and oops trace. This aids 709 debugging but costs about 70-100K of memory. 710 711config DEBUG_INFO 712 bool "Compile the kernel with debug info" 713 depends on DEBUG_KERNEL 714 help 715 If you say Y here the resulting kernel image will include 716 debugging info resulting in a larger kernel image. 717 This adds debug symbols to the kernel and modules (gcc -g), and 718 is needed if you intend to use kernel crashdump or binary object 719 tools like crash, kgdb, LKCD, gdb, etc on the kernel. 720 Say Y here only if you plan to debug the kernel. 721 722 If unsure, say N. 723 724config DEBUG_INFO_REDUCED 725 bool "Reduce debugging information" 726 depends on DEBUG_INFO 727 help 728 If you say Y here gcc is instructed to generate less debugging 729 information for structure types. This means that tools that 730 need full debugging information (like kgdb or systemtap) won't 731 be happy. But if you merely need debugging information to 732 resolve line numbers there is no loss. Advantage is that 733 build directory object sizes shrink dramatically over a full 734 DEBUG_INFO build and compile times are reduced too. 735 Only works with newer gcc versions. 736 737config DEBUG_VM 738 bool "Debug VM" 739 depends on DEBUG_KERNEL 740 help 741 Enable this to turn on extended checks in the virtual-memory system 742 that may impact performance. 743 744 If unsure, say N. 745 746config DEBUG_VIRTUAL 747 bool "Debug VM translations" 748 depends on DEBUG_KERNEL && X86 749 help 750 Enable some costly sanity checks in virtual to page code. This can 751 catch mistakes with virt_to_page() and friends. 752 753 If unsure, say N. 754 755config DEBUG_NOMMU_REGIONS 756 bool "Debug the global anon/private NOMMU mapping region tree" 757 depends on DEBUG_KERNEL && !MMU 758 help 759 This option causes the global tree of anonymous and private mapping 760 regions to be regularly checked for invalid topology. 761 762config DEBUG_WRITECOUNT 763 bool "Debug filesystem writers count" 764 depends on DEBUG_KERNEL 765 help 766 Enable this to catch wrong use of the writers count in struct 767 vfsmount. This will increase the size of each file struct by 768 32 bits. 769 770 If unsure, say N. 771 772config DEBUG_MEMORY_INIT 773 bool "Debug memory initialisation" if EXPERT 774 default !EXPERT 775 help 776 Enable this for additional checks during memory initialisation. 777 The sanity checks verify aspects of the VM such as the memory model 778 and other information provided by the architecture. Verbose 779 information will be printed at KERN_DEBUG loglevel depending 780 on the mminit_loglevel= command-line option. 781 782 If unsure, say Y 783 784config DEBUG_LIST 785 bool "Debug linked list manipulation" 786 depends on DEBUG_KERNEL 787 help 788 Enable this to turn on extended checks in the linked-list 789 walking routines. 790 791 If unsure, say N. 792 793config TEST_LIST_SORT 794 bool "Linked list sorting test" 795 depends on DEBUG_KERNEL 796 help 797 Enable this to turn on 'list_sort()' function test. This test is 798 executed only once during system boot, so affects only boot time. 799 800 If unsure, say N. 801 802config DEBUG_SG 803 bool "Debug SG table operations" 804 depends on DEBUG_KERNEL 805 help 806 Enable this to turn on checks on scatter-gather tables. This can 807 help find problems with drivers that do not properly initialize 808 their sg tables. 809 810 If unsure, say N. 811 812config DEBUG_NOTIFIERS 813 bool "Debug notifier call chains" 814 depends on DEBUG_KERNEL 815 help 816 Enable this to turn on sanity checking for notifier call chains. 817 This is most useful for kernel developers to make sure that 818 modules properly unregister themselves from notifier chains. 819 This is a relatively cheap check but if you care about maximum 820 performance, say N. 821 822config DEBUG_CREDENTIALS 823 bool "Debug credential management" 824 depends on DEBUG_KERNEL 825 help 826 Enable this to turn on some debug checking for credential 827 management. The additional code keeps track of the number of 828 pointers from task_structs to any given cred struct, and checks to 829 see that this number never exceeds the usage count of the cred 830 struct. 831 832 Furthermore, if SELinux is enabled, this also checks that the 833 security pointer in the cred struct is never seen to be invalid. 834 835 If unsure, say N. 836 837# 838# Select this config option from the architecture Kconfig, if it 839# is preferred to always offer frame pointers as a config 840# option on the architecture (regardless of KERNEL_DEBUG): 841# 842config ARCH_WANT_FRAME_POINTERS 843 bool 844 help 845 846config FRAME_POINTER 847 bool "Compile the kernel with frame pointers" 848 depends on DEBUG_KERNEL && \ 849 (CRIS || M68K || FRV || UML || \ 850 AVR32 || SUPERH || BLACKFIN || MN10300) || \ 851 ARCH_WANT_FRAME_POINTERS 852 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 853 help 854 If you say Y here the resulting kernel image will be slightly 855 larger and slower, but it gives very useful debugging information 856 in case of kernel bugs. (precise oopses/stacktraces/warnings) 857 858config BOOT_PRINTK_DELAY 859 bool "Delay each boot printk message by N milliseconds" 860 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY 861 help 862 This build option allows you to read kernel boot messages 863 by inserting a short delay after each one. The delay is 864 specified in milliseconds on the kernel command line, 865 using "boot_delay=N". 866 867 It is likely that you would also need to use "lpj=M" to preset 868 the "loops per jiffie" value. 869 See a previous boot log for the "lpj" value to use for your 870 system, and then set "lpj=M" before setting "boot_delay=N". 871 NOTE: Using this option may adversely affect SMP systems. 872 I.e., processors other than the first one may not boot up. 873 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect 874 what it believes to be lockup conditions. 875 876config RCU_TORTURE_TEST 877 tristate "torture tests for RCU" 878 depends on DEBUG_KERNEL 879 default n 880 help 881 This option provides a kernel module that runs torture tests 882 on the RCU infrastructure. The kernel module may be built 883 after the fact on the running kernel to be tested, if desired. 884 885 Say Y here if you want RCU torture tests to be built into 886 the kernel. 887 Say M if you want the RCU torture tests to build as a module. 888 Say N if you are unsure. 889 890config RCU_TORTURE_TEST_RUNNABLE 891 bool "torture tests for RCU runnable by default" 892 depends on RCU_TORTURE_TEST = y 893 default n 894 help 895 This option provides a way to build the RCU torture tests 896 directly into the kernel without them starting up at boot 897 time. You can use /proc/sys/kernel/rcutorture_runnable 898 to manually override this setting. This /proc file is 899 available only when the RCU torture tests have been built 900 into the kernel. 901 902 Say Y here if you want the RCU torture tests to start during 903 boot (you probably don't). 904 Say N here if you want the RCU torture tests to start only 905 after being manually enabled via /proc. 906 907config RCU_CPU_STALL_TIMEOUT 908 int "RCU CPU stall timeout in seconds" 909 depends on TREE_RCU || TREE_PREEMPT_RCU 910 range 3 300 911 default 60 912 help 913 If a given RCU grace period extends more than the specified 914 number of seconds, a CPU stall warning is printed. If the 915 RCU grace period persists, additional CPU stall warnings are 916 printed at more widely spaced intervals. 917 918config RCU_CPU_STALL_VERBOSE 919 bool "Print additional per-task information for RCU_CPU_STALL_DETECTOR" 920 depends on TREE_PREEMPT_RCU 921 default y 922 help 923 This option causes RCU to printk detailed per-task information 924 for any tasks that are stalling the current RCU grace period. 925 926 Say N if you are unsure. 927 928 Say Y if you want to enable such checks. 929 930config KPROBES_SANITY_TEST 931 bool "Kprobes sanity tests" 932 depends on DEBUG_KERNEL 933 depends on KPROBES 934 default n 935 help 936 This option provides for testing basic kprobes functionality on 937 boot. A sample kprobe, jprobe and kretprobe are inserted and 938 verified for functionality. 939 940 Say N if you are unsure. 941 942config BACKTRACE_SELF_TEST 943 tristate "Self test for the backtrace code" 944 depends on DEBUG_KERNEL 945 default n 946 help 947 This option provides a kernel module that can be used to test 948 the kernel stack backtrace code. This option is not useful 949 for distributions or general kernels, but only for kernel 950 developers working on architecture code. 951 952 Note that if you want to also test saved backtraces, you will 953 have to enable STACKTRACE as well. 954 955 Say N if you are unsure. 956 957config DEBUG_BLOCK_EXT_DEVT 958 bool "Force extended block device numbers and spread them" 959 depends on DEBUG_KERNEL 960 depends on BLOCK 961 default n 962 help 963 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 964 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 965 YOU ARE DOING. Distros, please enable this and fix whatever 966 is broken. 967 968 Conventionally, block device numbers are allocated from 969 predetermined contiguous area. However, extended block area 970 may introduce non-contiguous block device numbers. This 971 option forces most block device numbers to be allocated from 972 the extended space and spreads them to discover kernel or 973 userland code paths which assume predetermined contiguous 974 device number allocation. 975 976 Note that turning on this debug option shuffles all the 977 device numbers for all IDE and SCSI devices including libata 978 ones, so root partition specified using device number 979 directly (via rdev or root=MAJ:MIN) won't work anymore. 980 Textual device names (root=/dev/sdXn) will continue to work. 981 982 Say N if you are unsure. 983 984config DEBUG_FORCE_WEAK_PER_CPU 985 bool "Force weak per-cpu definitions" 986 depends on DEBUG_KERNEL 987 help 988 s390 and alpha require percpu variables in modules to be 989 defined weak to work around addressing range issue which 990 puts the following two restrictions on percpu variable 991 definitions. 992 993 1. percpu symbols must be unique whether static or not 994 2. percpu variables can't be defined inside a function 995 996 To ensure that generic code follows the above rules, this 997 option forces all percpu variables to be defined as weak. 998 999config DEBUG_PER_CPU_MAPS 1000 bool "Debug access to per_cpu maps"
1001 depends on DEBUG_KERNEL 1002 depends on SMP 1003 help 1004 Say Y to verify that the per_cpu map being accessed has 1005 been set up. This adds a fair amount of code to kernel memory 1006 and decreases performance. 1007 1008 Say N if unsure. 1009 1010config LKDTM 1011 tristate "Linux Kernel Dump Test Tool Module" 1012 depends on DEBUG_FS 1013 depends on BLOCK 1014 default n 1015 help 1016 This module enables testing of the different dumping mechanisms by 1017 inducing system failures at predefined crash points. 1018 If you don't need it: say N 1019 Choose M here to compile this code as a module. The module will be 1020 called lkdtm. 1021 1022 Documentation on how to use the module can be found in 1023 Documentation/fault-injection/provoke-crashes.txt 1024 1025config CPU_NOTIFIER_ERROR_INJECT 1026 tristate "CPU notifier error injection module" 1027 depends on HOTPLUG_CPU && DEBUG_KERNEL 1028 help 1029 This option provides a kernel module that can be used to test 1030 the error handling of the cpu notifiers 1031 1032 To compile this code as a module, choose M here: the module will 1033 be called cpu-notifier-error-inject. 1034 1035 If unsure, say N. 1036 1037config FAULT_INJECTION 1038 bool "Fault-injection framework" 1039 depends on DEBUG_KERNEL 1040 help 1041 Provide fault-injection framework. 1042 For more details, see Documentation/fault-injection/. 1043 1044config FAILSLAB 1045 bool "Fault-injection capability for kmalloc" 1046 depends on FAULT_INJECTION 1047 depends on SLAB || SLUB 1048 help 1049 Provide fault-injection capability for kmalloc. 1050 1051config FAIL_PAGE_ALLOC 1052 bool "Fault-injection capabilitiy for alloc_pages()" 1053 depends on FAULT_INJECTION 1054 help 1055 Provide fault-injection capability for alloc_pages(). 1056 1057config FAIL_MAKE_REQUEST 1058 bool "Fault-injection capability for disk IO" 1059 depends on FAULT_INJECTION && BLOCK 1060 help 1061 Provide fault-injection capability for disk IO. 1062 1063config FAIL_IO_TIMEOUT 1064 bool "Fault-injection capability for faking disk interrupts" 1065 depends on FAULT_INJECTION && BLOCK 1066 help 1067 Provide fault-injection capability on end IO handling. This 1068 will make the block layer "forget" an interrupt as configured, 1069 thus exercising the error handling. 1070 1071 Only works with drivers that use the generic timeout handling, 1072 for others it wont do anything. 1073 1074config FAIL_MMC_REQUEST 1075 bool "Fault-injection capability for MMC IO" 1076 select DEBUG_FS 1077 depends on FAULT_INJECTION && MMC 1078 help 1079 Provide fault-injection capability for MMC IO. 1080 This will make the mmc core return data errors. This is 1081 useful to test the error handling in the mmc block device 1082 and to test how the mmc host driver handles retries from 1083 the block device. 1084 1085config FAULT_INJECTION_DEBUG_FS 1086 bool "Debugfs entries for fault-injection capabilities" 1087 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1088 help 1089 Enable configuration of fault-injection capabilities via debugfs. 1090 1091config FAULT_INJECTION_STACKTRACE_FILTER 1092 bool "stacktrace filter for fault-injection capabilities" 1093 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1094 depends on !X86_64 1095 select STACKTRACE 1096 select FRAME_POINTER if !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND 1097 help 1098 Provide stacktrace filter for fault-injection capabilities 1099 1100config LATENCYTOP 1101 bool "Latency measuring infrastructure" 1102 depends on HAVE_LATENCYTOP_SUPPORT 1103 depends on DEBUG_KERNEL 1104 depends on STACKTRACE_SUPPORT 1105 depends on PROC_FS 1106 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND 1107 select KALLSYMS 1108 select KALLSYMS_ALL 1109 select STACKTRACE 1110 select SCHEDSTATS 1111 select SCHED_DEBUG 1112 help 1113 Enable this option if you want to use the LatencyTOP tool 1114 to find out which userspace is blocking on what kernel operations. 1115 1116config SYSCTL_SYSCALL_CHECK 1117 bool "Sysctl checks" 1118 depends on SYSCTL 1119 ---help--- 1120 sys_sysctl uses binary paths that have been found challenging 1121 to properly maintain and use. This enables checks that help 1122 you to keep things correct. 1123 1124source mm/Kconfig.debug 1125source kernel/trace/Kconfig 1126 1127config PROVIDE_OHCI1394_DMA_INIT 1128 bool "Remote debugging over FireWire early on boot" 1129 depends on PCI && X86 1130 help 1131 If you want to debug problems which hang or crash the kernel early 1132 on boot and the crashing machine has a FireWire port, you can use 1133 this feature to remotely access the memory of the crashed machine 1134 over FireWire. This employs remote DMA as part of the OHCI1394 1135 specification which is now the standard for FireWire controllers. 1136 1137 With remote DMA, you can monitor the printk buffer remotely using 1138 firescope and access all memory below 4GB using fireproxy from gdb. 1139 Even controlling a kernel debugger is possible using remote DMA. 1140 1141 Usage: 1142 1143 If ohci1394_dma=early is used as boot parameter, it will initialize 1144 all OHCI1394 controllers which are found in the PCI config space. 1145 1146 As all changes to the FireWire bus such as enabling and disabling 1147 devices cause a bus reset and thereby disable remote DMA for all 1148 devices, be sure to have the cable plugged and FireWire enabled on 1149 the debugging host before booting the debug target for debugging. 1150 1151 This code (~1k) is freed after boot. By then, the firewire stack 1152 in charge of the OHCI-1394 controllers should be used instead. 1153 1154 See Documentation/debugging-via-ohci1394.txt for more information. 1155 1156config FIREWIRE_OHCI_REMOTE_DMA 1157 bool "Remote debugging over FireWire with firewire-ohci" 1158 depends on FIREWIRE_OHCI 1159 help 1160 This option lets you use the FireWire bus for remote debugging 1161 with help of the firewire-ohci driver. It enables unfiltered 1162 remote DMA in firewire-ohci. 1163 See Documentation/debugging-via-ohci1394.txt for more information. 1164 1165 If unsure, say N. 1166 1167config BUILD_DOCSRC 1168 bool "Build targets in Documentation/ tree" 1169 depends on HEADERS_CHECK 1170 help 1171 This option attempts to build objects from the source files in the 1172 kernel Documentation/ tree. 1173 1174 Say N if you are unsure. 1175 1176config DYNAMIC_DEBUG 1177 bool "Enable dynamic printk() support" 1178 default n 1179 depends on PRINTK 1180 depends on DEBUG_FS 1181 help 1182 1183 Compiles debug level messages into the kernel, which would not 1184 otherwise be available at runtime. These messages can then be 1185 enabled/disabled based on various levels of scope - per source file, 1186 function, module, format string, and line number. This mechanism 1187 implicitly enables all pr_debug() and dev_dbg() calls. The impact of 1188 this compile option is a larger kernel text size of about 2%. 1189 1190 Usage: 1191 1192 Dynamic debugging is controlled via the 'dynamic_debug/control' file, 1193 which is contained in the 'debugfs' filesystem. Thus, the debugfs 1194 filesystem must first be mounted before making use of this feature. 1195 We refer the control file as: <debugfs>/dynamic_debug/control. This 1196 file contains a list of the debug statements that can be enabled. The 1197 format for each line of the file is: 1198 1199 filename:lineno [module]function flags format 1200 1201 filename : source file of the debug statement 1202 lineno : line number of the debug statement 1203 module : module that contains the debug statement 1204 function : function that contains the debug statement 1205 flags : 'p' means the line is turned 'on' for printing 1206 format : the format used for the debug statement 1207 1208 From a live system: 1209 1210 nullarbor:~ # cat <debugfs>/dynamic_debug/control 1211 # filename:lineno [module]function flags format 1212 fs/aio.c:222 [aio]__put_ioctx - "__put_ioctx:\040freeing\040%p\012" 1213 fs/aio.c:248 [aio]ioctx_alloc - "ENOMEM:\040nr_events\040too\040high\012" 1214 fs/aio.c:1770 [aio]sys_io_cancel - "calling\040cancel\012" 1215 1216 Example usage: 1217 1218 // enable the message at line 1603 of file svcsock.c 1219 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > 1220 <debugfs>/dynamic_debug/control 1221 1222 // enable all the messages in file svcsock.c 1223 nullarbor:~ # echo -n 'file svcsock.c +p' > 1224 <debugfs>/dynamic_debug/control 1225 1226 // enable all the messages in the NFS server module 1227 nullarbor:~ # echo -n 'module nfsd +p' > 1228 <debugfs>/dynamic_debug/control 1229 1230 // enable all 12 messages in the function svc_process() 1231 nullarbor:~ # echo -n 'func svc_process +p' > 1232 <debugfs>/dynamic_debug/control 1233 1234 // disable all 12 messages in the function svc_process() 1235 nullarbor:~ # echo -n 'func svc_process -p' > 1236 <debugfs>/dynamic_debug/control 1237 1238 See Documentation/dynamic-debug-howto.txt for additional information. 1239 1240config DMA_API_DEBUG 1241 bool "Enable debugging of DMA-API usage" 1242 depends on HAVE_DMA_API_DEBUG 1243 help 1244 Enable this option to debug the use of the DMA API by device drivers. 1245 With this option you will be able to detect common bugs in device 1246 drivers like double-freeing of DMA mappings or freeing mappings that 1247 were never allocated. 1248 This option causes a performance degredation. Use only if you want 1249 to debug device drivers. If unsure, say N. 1250 1251config ATOMIC64_SELFTEST 1252 bool "Perform an atomic64_t self-test at boot" 1253 help 1254 Enable this option to test the atomic64_t functions at boot. 1255 1256 If unsure, say N. 1257 1258config ASYNC_RAID6_TEST 1259 tristate "Self test for hardware accelerated raid6 recovery" 1260 depends on ASYNC_RAID6_RECOV 1261 select ASYNC_MEMCPY 1262 ---help--- 1263 This is a one-shot self test that permutes through the 1264 recovery of all the possible two disk failure scenarios for a 1265 N-disk array. Recovery is performed with the asynchronous 1266 raid6 recovery routines, and will optionally use an offload 1267 engine if one is available. 1268 1269 If unsure, say N. 1270 1271source "samples/Kconfig" 1272 1273source "lib/Kconfig.kgdb" 1274 1275source "lib/Kconfig.kmemcheck" 1276 1277config TEST_KSTRTOX 1278 tristate "Test kstrto*() family of functions at runtime" 1279