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/admin-guide/kernel-parameters.rst 17 18config CONSOLE_LOGLEVEL_DEFAULT 19 int "Default console loglevel (1-15)" 20 range 1 15 21 default "7" 22 help 23 Default loglevel to determine what will be printed on the console. 24 25 Setting a default here is equivalent to passing in loglevel=<x> in 26 the kernel bootargs. loglevel=<x> continues to override whatever 27 value is specified here as well. 28 29 Note: This does not affect the log level of un-prefixed printk() 30 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT 31 option. 32 33config MESSAGE_LOGLEVEL_DEFAULT 34 int "Default message log level (1-7)" 35 range 1 7 36 default "4" 37 help 38 Default log level for printk statements with no specified priority. 39 40 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks 41 that are auditing their logs closely may want to set it to a lower 42 priority. 43 44 Note: This does not affect what message level gets printed on the console 45 by default. To change that, use loglevel=<x> in the kernel bootargs, 46 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value. 47 48config BOOT_PRINTK_DELAY 49 bool "Delay each boot printk message by N milliseconds" 50 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY 51 help 52 This build option allows you to read kernel boot messages 53 by inserting a short delay after each one. The delay is 54 specified in milliseconds on the kernel command line, 55 using "boot_delay=N". 56 57 It is likely that you would also need to use "lpj=M" to preset 58 the "loops per jiffie" value. 59 See a previous boot log for the "lpj" value to use for your 60 system, and then set "lpj=M" before setting "boot_delay=N". 61 NOTE: Using this option may adversely affect SMP systems. 62 I.e., processors other than the first one may not boot up. 63 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect 64 what it believes to be lockup conditions. 65 66config DYNAMIC_DEBUG 67 bool "Enable dynamic printk() support" 68 default n 69 depends on PRINTK 70 depends on DEBUG_FS 71 help 72 73 Compiles debug level messages into the kernel, which would not 74 otherwise be available at runtime. These messages can then be 75 enabled/disabled based on various levels of scope - per source file, 76 function, module, format string, and line number. This mechanism 77 implicitly compiles in all pr_debug() and dev_dbg() calls, which 78 enlarges the kernel text size by about 2%. 79 80 If a source file is compiled with DEBUG flag set, any 81 pr_debug() calls in it are enabled by default, but can be 82 disabled at runtime as below. Note that DEBUG flag is 83 turned on by many CONFIG_*DEBUG* options. 84 85 Usage: 86 87 Dynamic debugging is controlled via the 'dynamic_debug/control' file, 88 which is contained in the 'debugfs' filesystem. Thus, the debugfs 89 filesystem must first be mounted before making use of this feature. 90 We refer the control file as: <debugfs>/dynamic_debug/control. This 91 file contains a list of the debug statements that can be enabled. The 92 format for each line of the file is: 93 94 filename:lineno [module]function flags format 95 96 filename : source file of the debug statement 97 lineno : line number of the debug statement 98 module : module that contains the debug statement 99 function : function that contains the debug statement 100 flags : '=p' means the line is turned 'on' for printing 101 format : the format used for the debug statement 102 103 From a live system: 104 105 nullarbor:~ # cat <debugfs>/dynamic_debug/control 106 # filename:lineno [module]function flags format 107 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012" 108 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012" 109 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012" 110 111 Example usage: 112 113 // enable the message at line 1603 of file svcsock.c 114 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > 115 <debugfs>/dynamic_debug/control 116 117 // enable all the messages in file svcsock.c 118 nullarbor:~ # echo -n 'file svcsock.c +p' > 119 <debugfs>/dynamic_debug/control 120 121 // enable all the messages in the NFS server module 122 nullarbor:~ # echo -n 'module nfsd +p' > 123 <debugfs>/dynamic_debug/control 124 125 // enable all 12 messages in the function svc_process() 126 nullarbor:~ # echo -n 'func svc_process +p' > 127 <debugfs>/dynamic_debug/control 128 129 // disable all 12 messages in the function svc_process() 130 nullarbor:~ # echo -n 'func svc_process -p' > 131 <debugfs>/dynamic_debug/control 132 133 See Documentation/admin-guide/dynamic-debug-howto.rst for additional 134 information. 135 136endmenu # "printk and dmesg options" 137 138menu "Compile-time checks and compiler options" 139 140config DEBUG_INFO 141 bool "Compile the kernel with debug info" 142 depends on DEBUG_KERNEL && !COMPILE_TEST 143 help 144 If you say Y here the resulting kernel image will include 145 debugging info resulting in a larger kernel image. 146 This adds debug symbols to the kernel and modules (gcc -g), and 147 is needed if you intend to use kernel crashdump or binary object 148 tools like crash, kgdb, LKCD, gdb, etc on the kernel. 149 Say Y here only if you plan to debug the kernel. 150 151 If unsure, say N. 152 153config DEBUG_INFO_REDUCED 154 bool "Reduce debugging information" 155 depends on DEBUG_INFO 156 help 157 If you say Y here gcc is instructed to generate less debugging 158 information for structure types. This means that tools that 159 need full debugging information (like kgdb or systemtap) won't 160 be happy. But if you merely need debugging information to 161 resolve line numbers there is no loss. Advantage is that 162 build directory object sizes shrink dramatically over a full 163 DEBUG_INFO build and compile times are reduced too. 164 Only works with newer gcc versions. 165 166config DEBUG_INFO_SPLIT 167 bool "Produce split debuginfo in .dwo files" 168 depends on DEBUG_INFO && !FRV 169 help 170 Generate debug info into separate .dwo files. This significantly 171 reduces the build directory size for builds with DEBUG_INFO, 172 because it stores the information only once on disk in .dwo 173 files instead of multiple times in object files and executables. 174 In addition the debug information is also compressed. 175 176 Requires recent gcc (4.7+) and recent gdb/binutils. 177 Any tool that packages or reads debug information would need 178 to know about the .dwo files and include them. 179 Incompatible with older versions of ccache. 180 181config DEBUG_INFO_DWARF4 182 bool "Generate dwarf4 debuginfo" 183 depends on DEBUG_INFO 184 help 185 Generate dwarf4 debug info. This requires recent versions 186 of gcc and gdb. It makes the debug information larger. 187 But it significantly improves the success of resolving 188 variables in gdb on optimized code. 189 190config GDB_SCRIPTS 191 bool "Provide GDB scripts for kernel debugging" 192 depends on DEBUG_INFO 193 help 194 This creates the required links to GDB helper scripts in the 195 build directory. If you load vmlinux into gdb, the helper 196 scripts will be automatically imported by gdb as well, and 197 additional functions are available to analyze a Linux kernel 198 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst 199 for further details. 200 201config ENABLE_WARN_DEPRECATED 202 bool "Enable __deprecated logic" 203 default y 204 help 205 Enable the __deprecated logic in the kernel build. 206 Disable this to suppress the "warning: 'foo' is deprecated 207 (declared at kernel/power/somefile.c:1234)" messages. 208 209config ENABLE_MUST_CHECK 210 bool "Enable __must_check logic" 211 default y 212 help 213 Enable the __must_check logic in the kernel build. Disable this to 214 suppress the "warning: ignoring return value of 'foo', declared with 215 attribute warn_unused_result" messages. 216 217config FRAME_WARN 218 int "Warn for stack frames larger than (needs gcc 4.4)" 219 range 0 8192 220 default 0 if KASAN 221 default 2048 if GCC_PLUGIN_LATENT_ENTROPY 222 default 1280 if (!64BIT && PARISC) 223 default 1024 if (!64BIT && !PARISC) 224 default 2048 if 64BIT 225 help 226 Tell gcc to warn at build time for stack frames larger than this. 227 Setting this too low will cause a lot of warnings. 228 Setting it to 0 disables the warning. 229 Requires gcc 4.4 230 231config STRIP_ASM_SYMS 232 bool "Strip assembler-generated symbols during link" 233 default n 234 help 235 Strip internal assembler-generated symbols during a link (symbols 236 that look like '.Lxxx') so they don't pollute the output of 237 get_wchan() and suchlike. 238 239config READABLE_ASM 240 bool "Generate readable assembler code" 241 depends on DEBUG_KERNEL 242 help 243 Disable some compiler optimizations that tend to generate human unreadable 244 assembler output. This may make the kernel slightly slower, but it helps 245 to keep kernel developers who have to stare a lot at assembler listings 246 sane. 247 248config UNUSED_SYMBOLS 249 bool "Enable unused/obsolete exported symbols" 250 default y if X86 251 help 252 Unused but exported symbols make the kernel needlessly bigger. For 253 that reason most of these unused exports will soon be removed. This 254 option is provided temporarily to provide a transition period in case 255 some external kernel module needs one of these symbols anyway. If you 256 encounter such a case in your module, consider if you are actually 257 using the right API. (rationale: since nobody in the kernel is using 258 this in a module, there is a pretty good chance it's actually the 259 wrong interface to use). If you really need the symbol, please send a 260 mail to the linux kernel mailing list mentioning the symbol and why 261 you really need it, and what the merge plan to the mainline kernel for 262 your module is. 263 264config PAGE_OWNER 265 bool "Track page owner" 266 depends on DEBUG_KERNEL && STACKTRACE_SUPPORT 267 select DEBUG_FS 268 select STACKTRACE 269 select STACKDEPOT 270 select PAGE_EXTENSION 271 help 272 This keeps track of what call chain is the owner of a page, may 273 help to find bare alloc_page(s) leaks. Even if you include this 274 feature on your build, it is disabled in default. You should pass 275 "page_owner=on" to boot parameter in order to enable it. Eats 276 a fair amount of memory if enabled. See tools/vm/page_owner_sort.c 277 for user-space helper. 278 279 If unsure, say N. 280 281config DEBUG_FS 282 bool "Debug Filesystem" 283 help 284 debugfs is a virtual file system that kernel developers use to put 285 debugging files into. Enable this option to be able to read and 286 write to these files. 287 288 For detailed documentation on the debugfs API, see 289 Documentation/filesystems/. 290 291 If unsure, say N. 292 293config HEADERS_CHECK 294 bool "Run 'make headers_check' when building vmlinux" 295 depends on !UML 296 help 297 This option will extract the user-visible kernel headers whenever 298 building the kernel, and will run basic sanity checks on them to 299 ensure that exported files do not attempt to include files which 300 were not exported, etc. 301 302 If you're making modifications to header files which are 303 relevant for userspace, say 'Y', and check the headers 304 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in 305 your build tree), to make sure they're suitable. 306 307config DEBUG_SECTION_MISMATCH 308 bool "Enable full Section mismatch analysis" 309 help 310 The section mismatch analysis checks if there are illegal 311 references from one section to another section. 312 During linktime or runtime, some sections are dropped; 313 any use of code/data previously in these sections would 314 most likely result in an oops. 315 In the code, functions and variables are annotated with 316 __init,, etc. (see the full list in include/linux/init.h), 317 which results in the code/data being placed in specific sections. 318 The section mismatch analysis is always performed after a full 319 kernel build, and enabling this option causes the following 320 additional steps to occur: 321 - Add the option -fno-inline-functions-called-once to gcc commands. 322 When inlining a function annotated with __init in a non-init 323 function, we would lose the section information and thus 324 the analysis would not catch the illegal reference. 325 This option tells gcc to inline less (but it does result in 326 a larger kernel). 327 - Run the section mismatch analysis for each module/built-in.o file. 328 When we run the section mismatch analysis on vmlinux.o, we 329 lose valuable information about where the mismatch was 330 introduced. 331 Running the analysis for each module/built-in.o file 332 tells where the mismatch happens much closer to the 333 source. The drawback is that the same mismatch is 334 reported at least twice. 335 - Enable verbose reporting from modpost in order to help resolve 336 the section mismatches that are reported. 337 338config SECTION_MISMATCH_WARN_ONLY 339 bool "Make section mismatch errors non-fatal" 340 default y 341 help 342 If you say N here, the build process will fail if there are any 343 section mismatch, instead of just throwing warnings. 344 345 If unsure, say Y. 346 347# 348# Select this config option from the architecture Kconfig, if it 349# is preferred to always offer frame pointers as a config 350# option on the architecture (regardless of KERNEL_DEBUG): 351# 352config ARCH_WANT_FRAME_POINTERS 353 bool 354 help 355 356config FRAME_POINTER 357 bool "Compile the kernel with frame pointers" 358 depends on DEBUG_KERNEL && \ 359 (CRIS || M68K || FRV || UML || \ 360 SUPERH || BLACKFIN || MN10300 || METAG) || \ 361 ARCH_WANT_FRAME_POINTERS 362 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 363 help 364 If you say Y here the resulting kernel image will be slightly 365 larger and slower, but it gives very useful debugging information 366 in case of kernel bugs. (precise oopses/stacktraces/warnings) 367 368config STACK_VALIDATION 369 bool "Compile-time stack metadata validation" 370 depends on HAVE_STACK_VALIDATION 371 default n 372 help 373 Add compile-time checks to validate stack metadata, including frame 374 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure 375 that runtime stack traces are more reliable. 376 377 This is also a prerequisite for generation of ORC unwind data, which 378 is needed for CONFIG_UNWINDER_ORC. 379 380 For more information, see 381 tools/objtool/Documentation/stack-validation.txt. 382 383config DEBUG_FORCE_WEAK_PER_CPU 384 bool "Force weak per-cpu definitions" 385 depends on DEBUG_KERNEL 386 help 387 s390 and alpha require percpu variables in modules to be 388 defined weak to work around addressing range issue which 389 puts the following two restrictions on percpu variable 390 definitions. 391 392 1. percpu symbols must be unique whether static or not 393 2. percpu variables can't be defined inside a function 394 395 To ensure that generic code follows the above rules, this 396 option forces all percpu variables to be defined as weak. 397 398endmenu # "Compiler options" 399 400config MAGIC_SYSRQ 401 bool "Magic SysRq key" 402 depends on !UML 403 help 404 If you say Y here, you will have some control over the system even 405 if the system crashes for example during kernel debugging (e.g., you 406 will be able to flush the buffer cache to disk, reboot the system 407 immediately or dump some status information). This is accomplished 408 by pressing various keys while holding SysRq (Alt+PrintScreen). It 409 also works on a serial console (on PC hardware at least), if you 410 send a BREAK and then within 5 seconds a command keypress. The 411 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 412 Don't say Y unless you really know what this hack does. 413 414config MAGIC_SYSRQ_DEFAULT_ENABLE 415 hex "Enable magic SysRq key functions by default" 416 depends on MAGIC_SYSRQ 417 default 0x1 418 help 419 Specifies which SysRq key functions are enabled by default. 420 This may be set to 1 or 0 to enable or disable them all, or 421 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 422 423config MAGIC_SYSRQ_SERIAL 424 bool "Enable magic SysRq key over serial" 425 depends on MAGIC_SYSRQ 426 default y 427 help 428 Many embedded boards have a disconnected TTL level serial which can 429 generate some garbage that can lead to spurious false sysrq detects. 430 This option allows you to decide whether you want to enable the 431 magic SysRq key. 432 433config DEBUG_KERNEL 434 bool "Kernel debugging" 435 help 436 Say Y here if you are developing drivers or trying to debug and 437 identify kernel problems. 438 439menu "Memory Debugging" 440 441source mm/Kconfig.debug 442 443config DEBUG_OBJECTS 444 bool "Debug object operations" 445 depends on DEBUG_KERNEL 446 help 447 If you say Y here, additional code will be inserted into the 448 kernel to track the life time of various objects and validate 449 the operations on those objects. 450 451config DEBUG_OBJECTS_SELFTEST 452 bool "Debug objects selftest" 453 depends on DEBUG_OBJECTS 454 help 455 This enables the selftest of the object debug code. 456 457config DEBUG_OBJECTS_FREE 458 bool "Debug objects in freed memory" 459 depends on DEBUG_OBJECTS 460 help 461 This enables checks whether a k/v free operation frees an area 462 which contains an object which has not been deactivated 463 properly. This can make kmalloc/kfree-intensive workloads 464 much slower. 465 466config DEBUG_OBJECTS_TIMERS 467 bool "Debug timer objects" 468 depends on DEBUG_OBJECTS 469 help 470 If you say Y here, additional code will be inserted into the 471 timer routines to track the life time of timer objects and 472 validate the timer operations. 473 474config DEBUG_OBJECTS_WORK 475 bool "Debug work objects" 476 depends on DEBUG_OBJECTS 477 help 478 If you say Y here, additional code will be inserted into the 479 work queue routines to track the life time of work objects and 480 validate the work operations. 481 482config DEBUG_OBJECTS_RCU_HEAD 483 bool "Debug RCU callbacks objects" 484 depends on DEBUG_OBJECTS 485 help 486 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 487 488config DEBUG_OBJECTS_PERCPU_COUNTER 489 bool "Debug percpu counter objects" 490 depends on DEBUG_OBJECTS 491 help 492 If you say Y here, additional code will be inserted into the 493 percpu counter routines to track the life time of percpu counter 494 objects and validate the percpu counter operations. 495 496config DEBUG_OBJECTS_ENABLE_DEFAULT 497 int "debug_objects bootup default value (0-1)" 498 range 0 1 499 default "1" 500 depends on DEBUG_OBJECTS 501 help 502 Debug objects boot parameter default value 503 504config DEBUG_SLAB 505 bool "Debug slab memory allocations" 506 depends on DEBUG_KERNEL && SLAB 507 help 508 Say Y here to have the kernel do limited verification on memory 509 allocation as well as poisoning memory on free to catch use of freed 510 memory. This can make kmalloc/kfree-intensive workloads much slower. 511 512config DEBUG_SLAB_LEAK 513 bool "Memory leak debugging" 514 depends on DEBUG_SLAB 515 516config SLUB_DEBUG_ON 517 bool "SLUB debugging on by default" 518 depends on SLUB && SLUB_DEBUG 519 default n 520 help 521 Boot with debugging on by default. SLUB boots by default with 522 the runtime debug capabilities switched off. Enabling this is 523 equivalent to specifying the "slub_debug" parameter on boot. 524 There is no support for more fine grained debug control like 525 possible with slub_debug=xxx. SLUB debugging may be switched 526 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 527 "slub_debug=-". 528 529config SLUB_STATS 530 default n 531 bool "Enable SLUB performance statistics" 532 depends on SLUB && SYSFS 533 help 534 SLUB statistics are useful to debug SLUBs allocation behavior in 535 order find ways to optimize the allocator. This should never be 536 enabled for production use since keeping statistics slows down 537 the allocator by a few percentage points. The slabinfo command 538 supports the determination of the most active slabs to figure 539 out which slabs are relevant to a particular load. 540 Try running: slabinfo -DA 541 542config HAVE_DEBUG_KMEMLEAK 543 bool 544 545config DEBUG_KMEMLEAK 546 bool "Kernel memory leak detector" 547 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 548 select DEBUG_FS 549 select STACKTRACE if STACKTRACE_SUPPORT 550 select KALLSYMS 551 select CRC32 552 help 553 Say Y here if you want to enable the memory leak 554 detector. The memory allocation/freeing is traced in a way 555 similar to the Boehm's conservative garbage collector, the 556 difference being that the orphan objects are not freed but 557 only shown in /sys/kernel/debug/kmemleak. Enabling this 558 feature will introduce an overhead to memory 559 allocations. See Documentation/dev-tools/kmemleak.rst for more 560 details. 561 562 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 563 of finding leaks due to the slab objects poisoning. 564 565 In order to access the kmemleak file, debugfs needs to be 566 mounted (usually at /sys/kernel/debug). 567 568config DEBUG_KMEMLEAK_EARLY_LOG_SIZE 569 int "Maximum kmemleak early log entries" 570 depends on DEBUG_KMEMLEAK 571 range 200 40000 572 default 400 573 help 574 Kmemleak must track all the memory allocations to avoid 575 reporting false positives. Since memory may be allocated or 576 freed before kmemleak is initialised, an early log buffer is 577 used to store these actions. If kmemleak reports "early log 578 buffer exceeded", please increase this value. 579 580config DEBUG_KMEMLEAK_TEST 581 tristate "Simple test for the kernel memory leak detector" 582 depends on DEBUG_KMEMLEAK && m 583 help 584 This option enables a module that explicitly leaks memory. 585 586 If unsure, say N. 587 588config DEBUG_KMEMLEAK_DEFAULT_OFF 589 bool "Default kmemleak to off" 590 depends on DEBUG_KMEMLEAK 591 help 592 Say Y here to disable kmemleak by default. It can then be enabled 593 on the command line via kmemleak=on. 594 595config DEBUG_STACK_USAGE 596 bool "Stack utilization instrumentation" 597 depends on DEBUG_KERNEL && !IA64 598 help 599 Enables the display of the minimum amount of free stack which each 600 task has ever had available in the sysrq-T and sysrq-P debug output. 601 602 This option will slow down process creation somewhat. 603 604config DEBUG_VM 605 bool "Debug VM" 606 depends on DEBUG_KERNEL 607 help 608 Enable this to turn on extended checks in the virtual-memory system 609 that may impact performance. 610 611 If unsure, say N. 612 613config DEBUG_VM_VMACACHE 614 bool "Debug VMA caching" 615 depends on DEBUG_VM 616 help 617 Enable this to turn on VMA caching debug information. Doing so 618 can cause significant overhead, so only enable it in non-production 619 environments. 620 621 If unsure, say N. 622 623config DEBUG_VM_RB 624 bool "Debug VM red-black trees" 625 depends on DEBUG_VM 626 help 627 Enable VM red-black tree debugging information and extra validations. 628 629 If unsure, say N. 630 631config DEBUG_VM_PGFLAGS 632 bool "Debug page-flags operations" 633 depends on DEBUG_VM 634 help 635 Enables extra validation on page flags operations. 636 637 If unsure, say N. 638 639config ARCH_HAS_DEBUG_VIRTUAL 640 bool 641 642config DEBUG_VIRTUAL 643 bool "Debug VM translations" 644 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 645 help 646 Enable some costly sanity checks in virtual to page code. This can 647 catch mistakes with virt_to_page() and friends. 648 649 If unsure, say N. 650 651config DEBUG_NOMMU_REGIONS 652 bool "Debug the global anon/private NOMMU mapping region tree" 653 depends on DEBUG_KERNEL && !MMU 654 help 655 This option causes the global tree of anonymous and private mapping 656 regions to be regularly checked for invalid topology. 657 658config DEBUG_MEMORY_INIT 659 bool "Debug memory initialisation" if EXPERT 660 default !EXPERT 661 help 662 Enable this for additional checks during memory initialisation. 663 The sanity checks verify aspects of the VM such as the memory model 664 and other information provided by the architecture. Verbose 665 information will be printed at KERN_DEBUG loglevel depending 666 on the mminit_loglevel= command-line option. 667 668 If unsure, say Y 669 670config MEMORY_NOTIFIER_ERROR_INJECT 671 tristate "Memory hotplug notifier error injection module" 672 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 673 help 674 This option provides the ability to inject artificial errors to 675 memory hotplug notifier chain callbacks. It is controlled through 676 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 677 678 If the notifier call chain should be failed with some events 679 notified, write the error code to "actions/<notifier event>/error". 680 681 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 682 683 # cd /sys/kernel/debug/notifier-error-inject/memory 684 # echo -12 > actions/MEM_GOING_OFFLINE/error 685 # echo offline > /sys/devices/system/memory/memoryXXX/state 686 bash: echo: write error: Cannot allocate memory 687 688 To compile this code as a module, choose M here: the module will 689 be called memory-notifier-error-inject. 690 691 If unsure, say N. 692 693config DEBUG_PER_CPU_MAPS 694 bool "Debug access to per_cpu maps" 695 depends on DEBUG_KERNEL 696 depends on SMP 697 help 698 Say Y to verify that the per_cpu map being accessed has 699 been set up. This adds a fair amount of code to kernel memory 700 and decreases performance. 701 702 Say N if unsure. 703 704config DEBUG_HIGHMEM 705 bool "Highmem debugging" 706 depends on DEBUG_KERNEL && HIGHMEM 707 help 708 This option enables additional error checking for high memory 709 systems. Disable for production systems. 710 711config HAVE_DEBUG_STACKOVERFLOW 712 bool 713 714config DEBUG_STACKOVERFLOW 715 bool "Check for stack overflows" 716 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 717 ---help--- 718 Say Y here if you want to check for overflows of kernel, IRQ 719 and exception stacks (if your architecture uses them). This 720 option will show detailed messages if free stack space drops 721 below a certain limit. 722 723 These kinds of bugs usually occur when call-chains in the 724 kernel get too deep, especially when interrupts are 725 involved. 726 727 Use this in cases where you see apparently random memory 728 corruption, especially if it appears in 'struct thread_info' 729 730 If in doubt, say "N". 731 732source "lib/Kconfig.kasan" 733 734endmenu # "Memory Debugging" 735 736config ARCH_HAS_KCOV 737 bool 738 help 739 KCOV does not have any arch-specific code, but currently it is enabled 740 only for x86_64. KCOV requires testing on other archs, and most likely 741 disabling of instrumentation for some early boot code. 742 743config KCOV 744 bool "Code coverage for fuzzing" 745 depends on ARCH_HAS_KCOV 746 select DEBUG_FS 747 select GCC_PLUGINS if !COMPILE_TEST 748 select GCC_PLUGIN_SANCOV if !COMPILE_TEST 749 help 750 KCOV exposes kernel code coverage information in a form suitable 751 for coverage-guided fuzzing (randomized testing). 752 753 If RANDOMIZE_BASE is enabled, PC values will not be stable across 754 different machines and across reboots. If you need stable PC values, 755 disable RANDOMIZE_BASE. 756 757 For more details, see Documentation/dev-tools/kcov.rst. 758 759config KCOV_INSTRUMENT_ALL 760 bool "Instrument all code by default" 761 depends on KCOV 762 default y if KCOV 763 help 764 If you are doing generic system call fuzzing (like e.g. syzkaller), 765 then you will want to instrument the whole kernel and you should 766 say y here. If you are doing more targeted fuzzing (like e.g. 767 filesystem fuzzing with AFL) then you will want to enable coverage 768 for more specific subsets of files, and should say n here. 769 770config DEBUG_SHIRQ 771 bool "Debug shared IRQ handlers" 772 depends on DEBUG_KERNEL 773 help 774 Enable this to generate a spurious interrupt as soon as a shared 775 interrupt handler is registered, and just before one is deregistered. 776 Drivers ought to be able to handle interrupts coming in at those 777 points; some don't and need to be caught. 778 779menu "Debug Lockups and Hangs" 780 781config LOCKUP_DETECTOR 782 bool 783 784config SOFTLOCKUP_DETECTOR 785 bool "Detect Soft Lockups" 786 depends on DEBUG_KERNEL && !S390 787 select LOCKUP_DETECTOR 788 help 789 Say Y here to enable the kernel to act as a watchdog to detect 790 soft lockups. 791 792 Softlockups are bugs that cause the kernel to loop in kernel 793 mode for more than 20 seconds, without giving other tasks a 794 chance to run. The current stack trace is displayed upon 795 detection and the system will stay locked up. 796 797config HARDLOCKUP_DETECTOR_PERF 798 bool 799 select SOFTLOCKUP_DETECTOR 800 801# 802# Enables a timestamp based low pass filter to compensate for perf based 803# hard lockup detection which runs too fast due to turbo modes. 804# 805config HARDLOCKUP_CHECK_TIMESTAMP 806 bool 807 808# 809# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard 810# lockup detector rather than the perf based detector. 811# 812config HARDLOCKUP_DETECTOR 813 bool "Detect Hard Lockups" 814 depends on DEBUG_KERNEL && !S390 815 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH 816 select LOCKUP_DETECTOR 817 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF 818 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH 819 help 820 Say Y here to enable the kernel to act as a watchdog to detect 821 hard lockups. 822 823 Hardlockups are bugs that cause the CPU to loop in kernel mode 824 for more than 10 seconds, without letting other interrupts have a 825 chance to run. The current stack trace is displayed upon detection 826 and the system will stay locked up. 827 828config BOOTPARAM_HARDLOCKUP_PANIC 829 bool "Panic (Reboot) On Hard Lockups" 830 depends on HARDLOCKUP_DETECTOR 831 help 832 Say Y here to enable the kernel to panic on "hard lockups", 833 which are bugs that cause the kernel to loop in kernel 834 mode with interrupts disabled for more than 10 seconds (configurable 835 using the watchdog_thresh sysctl). 836 837 Say N if unsure. 838 839config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 840 int 841 depends on HARDLOCKUP_DETECTOR 842 range 0 1 843 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 844 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 845 846config BOOTPARAM_SOFTLOCKUP_PANIC 847 bool "Panic (Reboot) On Soft Lockups" 848 depends on SOFTLOCKUP_DETECTOR 849 help 850 Say Y here to enable the kernel to panic on "soft lockups", 851 which are bugs that cause the kernel to loop in kernel 852 mode for more than 20 seconds (configurable using the watchdog_thresh 853 sysctl), without giving other tasks a chance to run. 854 855 The panic can be used in combination with panic_timeout, 856 to cause the system to reboot automatically after a 857 lockup has been detected. This feature is useful for 858 high-availability systems that have uptime guarantees and 859 where a lockup must be resolved ASAP. 860 861 Say N if unsure. 862 863config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 864 int 865 depends on SOFTLOCKUP_DETECTOR 866 range 0 1 867 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 868 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 869 870config DETECT_HUNG_TASK 871 bool "Detect Hung Tasks" 872 depends on DEBUG_KERNEL 873 default SOFTLOCKUP_DETECTOR 874 help 875 Say Y here to enable the kernel to detect "hung tasks", 876 which are bugs that cause the task to be stuck in 877 uninterruptible "D" state indefinitely. 878 879 When a hung task is detected, the kernel will print the 880 current stack trace (which you should report), but the 881 task will stay in uninterruptible state. If lockdep is 882 enabled then all held locks will also be reported. This 883 feature has negligible overhead. 884 885config DEFAULT_HUNG_TASK_TIMEOUT 886 int "Default timeout for hung task detection (in seconds)" 887 depends on DETECT_HUNG_TASK 888 default 120 889 help 890 This option controls the default timeout (in seconds) used 891 to determine when a task has become non-responsive and should 892 be considered hung. 893 894 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 895 sysctl or by writing a value to 896 /proc/sys/kernel/hung_task_timeout_secs. 897 898 A timeout of 0 disables the check. The default is two minutes. 899 Keeping the default should be fine in most cases. 900 901config BOOTPARAM_HUNG_TASK_PANIC 902 bool "Panic (Reboot) On Hung Tasks" 903 depends on DETECT_HUNG_TASK 904 help 905 Say Y here to enable the kernel to panic on "hung tasks", 906 which are bugs that cause the kernel to leave a task stuck 907 in uninterruptible "D" state. 908 909 The panic can be used in combination with panic_timeout, 910 to cause the system to reboot automatically after a 911 hung task has been detected. This feature is useful for 912 high-availability systems that have uptime guarantees and 913 where a hung tasks must be resolved ASAP. 914 915 Say N if unsure. 916 917config BOOTPARAM_HUNG_TASK_PANIC_VALUE 918 int 919 depends on DETECT_HUNG_TASK 920 range 0 1 921 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 922 default 1 if BOOTPARAM_HUNG_TASK_PANIC 923 924config WQ_WATCHDOG 925 bool "Detect Workqueue Stalls" 926 depends on DEBUG_KERNEL 927 help 928 Say Y here to enable stall detection on workqueues. If a 929 worker pool doesn't make forward progress on a pending work 930 item for over a given amount of time, 30s by default, a 931 warning message is printed along with dump of workqueue 932 state. This can be configured through kernel parameter 933 "workqueue.watchdog_thresh" and its sysfs counterpart. 934 935endmenu # "Debug lockups and hangs" 936 937config PANIC_ON_OOPS 938 bool "Panic on Oops" 939 help 940 Say Y here to enable the kernel to panic when it oopses. This 941 has the same effect as setting oops=panic on the kernel command 942 line. 943 944 This feature is useful to ensure that the kernel does not do 945 anything erroneous after an oops which could result in data 946 corruption or other issues. 947 948 Say N if unsure. 949 950config PANIC_ON_OOPS_VALUE 951 int 952 range 0 1 953 default 0 if !PANIC_ON_OOPS 954 default 1 if PANIC_ON_OOPS 955 956config PANIC_TIMEOUT 957 int "panic timeout" 958 default 0 959 help 960 Set the timeout value (in seconds) until a reboot occurs when the 961 the kernel panics. If n = 0, then we wait forever. A timeout 962 value n > 0 will wait n seconds before rebooting, while a timeout 963 value n < 0 will reboot immediately. 964 965config SCHED_DEBUG 966 bool "Collect scheduler debugging info" 967 depends on DEBUG_KERNEL && PROC_FS 968 default y 969 help 970 If you say Y here, the /proc/sched_debug file will be provided 971 that can help debug the scheduler. The runtime overhead of this 972 option is minimal. 973 974config SCHED_INFO 975 bool 976 default n 977 978config SCHEDSTATS 979 bool "Collect scheduler statistics" 980 depends on DEBUG_KERNEL && PROC_FS 981 select SCHED_INFO 982 help 983 If you say Y here, additional code will be inserted into the 984 scheduler and related routines to collect statistics about 985 scheduler behavior and provide them in /proc/schedstat. These 986 stats may be useful for both tuning and debugging the scheduler 987 If you aren't debugging the scheduler or trying to tune a specific 988 application, you can say N to avoid the very slight overhead 989 this adds. 990 991config SCHED_STACK_END_CHECK 992 bool "Detect stack corruption on calls to schedule()" 993 depends on DEBUG_KERNEL 994 default n 995 help 996 This option checks for a stack overrun on calls to schedule(). 997 If the stack end location is found to be over written always panic as 998 the content of the corrupted region can no longer be trusted. 999 This is to ensure no erroneous behaviour occurs which could result in 1000 data corruption or a sporadic crash at a later stage once the region 1001 is examined. The runtime overhead introduced is minimal. 1002 1003config DEBUG_TIMEKEEPING 1004 bool "Enable extra timekeeping sanity checking" 1005 help 1006 This option will enable additional timekeeping sanity checks 1007 which may be helpful when diagnosing issues where timekeeping 1008 problems are suspected. 1009 1010 This may include checks in the timekeeping hotpaths, so this 1011 option may have a (very small) performance impact to some 1012 workloads. 1013 1014 If unsure, say N. 1015 1016config DEBUG_PREEMPT 1017 bool "Debug preemptible kernel" 1018 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT 1019 default y 1020 help 1021 If you say Y here then the kernel will use a debug variant of the 1022 commonly used smp_processor_id() function and will print warnings 1023 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1024 will detect preemption count underflows. 1025 1026menu "Lock Debugging (spinlocks, mutexes, etc...)" 1027 1028config DEBUG_RT_MUTEXES 1029 bool "RT Mutex debugging, deadlock detection" 1030 depends on DEBUG_KERNEL && RT_MUTEXES 1031 help 1032 This allows rt mutex semantics violations and rt mutex related 1033 deadlocks (lockups) to be detected and reported automatically. 1034 1035config DEBUG_SPINLOCK 1036 bool "Spinlock and rw-lock debugging: basic checks" 1037 depends on DEBUG_KERNEL 1038 select UNINLINE_SPIN_UNLOCK 1039 help 1040 Say Y here and build SMP to catch missing spinlock initialization 1041 and certain other kinds of spinlock errors commonly made. This is 1042 best used in conjunction with the NMI watchdog so that spinlock 1043 deadlocks are also debuggable. 1044 1045config DEBUG_MUTEXES 1046 bool "Mutex debugging: basic checks" 1047 depends on DEBUG_KERNEL 1048 help 1049 This feature allows mutex semantics violations to be detected and 1050 reported. 1051 1052config DEBUG_WW_MUTEX_SLOWPATH 1053 bool "Wait/wound mutex debugging: Slowpath testing" 1054 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1055 select DEBUG_LOCK_ALLOC 1056 select DEBUG_SPINLOCK 1057 select DEBUG_MUTEXES 1058 help 1059 This feature enables slowpath testing for w/w mutex users by 1060 injecting additional -EDEADLK wound/backoff cases. Together with 1061 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1062 will test all possible w/w mutex interface abuse with the 1063 exception of simply not acquiring all the required locks. 1064 Note that this feature can introduce significant overhead, so 1065 it really should not be enabled in a production or distro kernel, 1066 even a debug kernel. If you are a driver writer, enable it. If 1067 you are a distro, do not. 1068 1069config DEBUG_LOCK_ALLOC 1070 bool "Lock debugging: detect incorrect freeing of live locks" 1071 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1072 select DEBUG_SPINLOCK 1073 select DEBUG_MUTEXES 1074 select DEBUG_RT_MUTEXES if RT_MUTEXES 1075 select LOCKDEP 1076 help 1077 This feature will check whether any held lock (spinlock, rwlock, 1078 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1079 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1080 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1081 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1082 held during task exit. 1083 1084config PROVE_LOCKING 1085 bool "Lock debugging: prove locking correctness" 1086 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1087 select LOCKDEP 1088 select DEBUG_SPINLOCK 1089 select DEBUG_MUTEXES 1090 select DEBUG_RT_MUTEXES if RT_MUTEXES 1091 select DEBUG_LOCK_ALLOC 1092 select LOCKDEP_CROSSRELEASE 1093 select LOCKDEP_COMPLETIONS 1094 select TRACE_IRQFLAGS 1095 default n 1096 help 1097 This feature enables the kernel to prove that all locking 1098 that occurs in the kernel runtime is mathematically 1099 correct: that under no circumstance could an arbitrary (and 1100 not yet triggered) combination of observed locking 1101 sequences (on an arbitrary number of CPUs, running an 1102 arbitrary number of tasks and interrupt contexts) cause a 1103 deadlock. 1104 1105 In short, this feature enables the kernel to report locking 1106 related deadlocks before they actually occur. 1107 1108 The proof does not depend on how hard and complex a 1109 deadlock scenario would be to trigger: how many 1110 participant CPUs, tasks and irq-contexts would be needed 1111 for it to trigger. The proof also does not depend on 1112 timing: if a race and a resulting deadlock is possible 1113 theoretically (no matter how unlikely the race scenario 1114 is), it will be proven so and will immediately be 1115 reported by the kernel (once the event is observed that 1116 makes the deadlock theoretically possible). 1117 1118 If a deadlock is impossible (i.e. the locking rules, as 1119 observed by the kernel, are mathematically correct), the 1120 kernel reports nothing. 1121 1122 NOTE: this feature can also be enabled for rwlocks, mutexes 1123 and rwsems - in which case all dependencies between these 1124 different locking variants are observed and mapped too, and 1125 the proof of observed correctness is also maintained for an 1126 arbitrary combination of these separate locking variants. 1127 1128 For more details, see Documentation/locking/lockdep-design.txt. 1129 1130config LOCKDEP 1131 bool 1132 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1133 select STACKTRACE 1134 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE && !ARC && !SCORE && !X86 1135 select KALLSYMS 1136 select KALLSYMS_ALL 1137 1138config LOCKDEP_SMALL 1139 bool 1140 1141config LOCK_STAT 1142 bool "Lock usage statistics" 1143 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1144 select LOCKDEP 1145 select DEBUG_SPINLOCK 1146 select DEBUG_MUTEXES 1147 select DEBUG_RT_MUTEXES if RT_MUTEXES 1148 select DEBUG_LOCK_ALLOC 1149 default n 1150 help 1151 This feature enables tracking lock contention points 1152 1153 For more details, see Documentation/locking/lockstat.txt 1154 1155 This also enables lock events required by "perf lock", 1156 subcommand of perf. 1157 If you want to use "perf lock", you also need to turn on 1158 CONFIG_EVENT_TRACING. 1159 1160 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1161 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1162 1163config LOCKDEP_CROSSRELEASE 1164 bool 1165 help 1166 This makes lockdep work for crosslock which is a lock allowed to 1167 be released in a different context from the acquisition context. 1168 Normally a lock must be released in the context acquiring the lock. 1169 However, relexing this constraint helps synchronization primitives 1170 such as page locks or completions can use the lock correctness 1171 detector, lockdep. 1172 1173config LOCKDEP_COMPLETIONS 1174 bool 1175 help 1176 A deadlock caused by wait_for_completion() and complete() can be 1177 detected by lockdep using crossrelease feature. 1178 1179config BOOTPARAM_LOCKDEP_CROSSRELEASE_FULLSTACK 1180 bool "Enable the boot parameter, crossrelease_fullstack" 1181 depends on LOCKDEP_CROSSRELEASE 1182 default n 1183 help 1184 The lockdep "cross-release" feature needs to record stack traces 1185 (of calling functions) for all acquisitions, for eventual later 1186 use during analysis. By default only a single caller is recorded, 1187 because the unwind operation can be very expensive with deeper 1188 stack chains. 1189 1190 However a boot parameter, crossrelease_fullstack, was 1191 introduced since sometimes deeper traces are required for full 1192 analysis. This option turns on the boot parameter. 1193 1194config DEBUG_LOCKDEP 1195 bool "Lock dependency engine debugging" 1196 depends on DEBUG_KERNEL && LOCKDEP 1197 help 1198 If you say Y here, the lock dependency engine will do 1199 additional runtime checks to debug itself, at the price 1200 of more runtime overhead. 1201 1202config DEBUG_ATOMIC_SLEEP 1203 bool "Sleep inside atomic section checking" 1204 select PREEMPT_COUNT 1205 depends on DEBUG_KERNEL 1206 help 1207 If you say Y here, various routines which may sleep will become very 1208 noisy if they are called inside atomic sections: when a spinlock is 1209 held, inside an rcu read side critical section, inside preempt disabled 1210 sections, inside an interrupt, etc... 1211 1212config DEBUG_LOCKING_API_SELFTESTS 1213 bool "Locking API boot-time self-tests" 1214 depends on DEBUG_KERNEL 1215 help 1216 Say Y here if you want the kernel to run a short self-test during 1217 bootup. The self-test checks whether common types of locking bugs 1218 are detected by debugging mechanisms or not. (if you disable 1219 lock debugging then those bugs wont be detected of course.) 1220 The following locking APIs are covered: spinlocks, rwlocks, 1221 mutexes and rwsems. 1222 1223config LOCK_TORTURE_TEST 1224 tristate "torture tests for locking" 1225 depends on DEBUG_KERNEL 1226 select TORTURE_TEST 1227 default n 1228 help 1229 This option provides a kernel module that runs torture tests 1230 on kernel locking primitives. The kernel module may be built 1231 after the fact on the running kernel to be tested, if desired. 1232 1233 Say Y here if you want kernel locking-primitive torture tests 1234 to be built into the kernel. 1235 Say M if you want these torture tests to build as a module. 1236 Say N if you are unsure. 1237 1238config WW_MUTEX_SELFTEST 1239 tristate "Wait/wound mutex selftests" 1240 help 1241 This option provides a kernel module that runs tests on the 1242 on the struct ww_mutex locking API. 1243 1244 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1245 with this test harness. 1246 1247 Say M if you want these self tests to build as a module. 1248 Say N if you are unsure. 1249 1250endmenu # lock debugging 1251 1252config TRACE_IRQFLAGS 1253 bool 1254 help 1255 Enables hooks to interrupt enabling and disabling for 1256 either tracing or lock debugging. 1257 1258config STACKTRACE 1259 bool "Stack backtrace support" 1260 depends on STACKTRACE_SUPPORT 1261 help 1262 This option causes the kernel to create a /proc/pid/stack for 1263 every process, showing its current stack trace. 1264 It is also used by various kernel debugging features that require 1265 stack trace generation. 1266 1267config WARN_ALL_UNSEEDED_RANDOM 1268 bool "Warn for all uses of unseeded randomness" 1269 default n 1270 help 1271 Some parts of the kernel contain bugs relating to their use of 1272 cryptographically secure random numbers before it's actually possible 1273 to generate those numbers securely. This setting ensures that these 1274 flaws don't go unnoticed, by enabling a message, should this ever 1275 occur. This will allow people with obscure setups to know when things 1276 are going wrong, so that they might contact developers about fixing 1277 it. 1278 1279 Unfortunately, on some models of some architectures getting 1280 a fully seeded CRNG is extremely difficult, and so this can 1281 result in dmesg getting spammed for a surprisingly long 1282 time. This is really bad from a security perspective, and 1283 so architecture maintainers really need to do what they can 1284 to get the CRNG seeded sooner after the system is booted. 1285 However, since users can not do anything actionble to 1286 address this, by default the kernel will issue only a single 1287 warning for the first use of unseeded randomness. 1288 1289 Say Y here if you want to receive warnings for all uses of 1290 unseeded randomness. This will be of use primarily for 1291 those developers interersted in improving the security of 1292 Linux kernels running on their architecture (or 1293 subarchitecture). 1294 1295config DEBUG_KOBJECT 1296 bool "kobject debugging" 1297 depends on DEBUG_KERNEL 1298 help 1299 If you say Y here, some extra kobject debugging messages will be sent 1300 to the syslog. 1301 1302config DEBUG_KOBJECT_RELEASE 1303 bool "kobject release debugging" 1304 depends on DEBUG_OBJECTS_TIMERS 1305 help 1306 kobjects are reference counted objects. This means that their 1307 last reference count put is not predictable, and the kobject can 1308 live on past the point at which a driver decides to drop it's 1309 initial reference to the kobject gained on allocation. An 1310 example of this would be a struct device which has just been 1311 unregistered. 1312 1313 However, some buggy drivers assume that after such an operation, 1314 the memory backing the kobject can be immediately freed. This 1315 goes completely against the principles of a refcounted object. 1316 1317 If you say Y here, the kernel will delay the release of kobjects 1318 on the last reference count to improve the visibility of this 1319 kind of kobject release bug. 1320 1321config HAVE_DEBUG_BUGVERBOSE 1322 bool 1323 1324config DEBUG_BUGVERBOSE 1325 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 1326 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE) 1327 default y 1328 help 1329 Say Y here to make BUG() panics output the file name and line number 1330 of the BUG call as well as the EIP and oops trace. This aids 1331 debugging but costs about 70-100K of memory. 1332 1333config DEBUG_LIST 1334 bool "Debug linked list manipulation" 1335 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION 1336 help 1337 Enable this to turn on extended checks in the linked-list 1338 walking routines. 1339 1340 If unsure, say N. 1341 1342config DEBUG_PI_LIST 1343 bool "Debug priority linked list manipulation" 1344 depends on DEBUG_KERNEL 1345 help 1346 Enable this to turn on extended checks in the priority-ordered 1347 linked-list (plist) walking routines. This checks the entire 1348 list multiple times during each manipulation. 1349 1350 If unsure, say N. 1351 1352config DEBUG_SG 1353 bool "Debug SG table operations" 1354 depends on DEBUG_KERNEL 1355 help 1356 Enable this to turn on checks on scatter-gather tables. This can 1357 help find problems with drivers that do not properly initialize 1358 their sg tables. 1359 1360 If unsure, say N. 1361 1362config DEBUG_NOTIFIERS 1363 bool "Debug notifier call chains" 1364 depends on DEBUG_KERNEL 1365 help 1366 Enable this to turn on sanity checking for notifier call chains. 1367 This is most useful for kernel developers to make sure that 1368 modules properly unregister themselves from notifier chains. 1369 This is a relatively cheap check but if you care about maximum 1370 performance, say N. 1371 1372config DEBUG_CREDENTIALS 1373 bool "Debug credential management" 1374 depends on DEBUG_KERNEL 1375 help 1376 Enable this to turn on some debug checking for credential 1377 management. The additional code keeps track of the number of 1378 pointers from task_structs to any given cred struct, and checks to 1379 see that this number never exceeds the usage count of the cred 1380 struct. 1381 1382 Furthermore, if SELinux is enabled, this also checks that the 1383 security pointer in the cred struct is never seen to be invalid. 1384 1385 If unsure, say N. 1386 1387source "kernel/rcu/Kconfig.debug" 1388 1389config DEBUG_WQ_FORCE_RR_CPU 1390 bool "Force round-robin CPU selection for unbound work items" 1391 depends on DEBUG_KERNEL 1392 default n 1393 help 1394 Workqueue used to implicitly guarantee that work items queued 1395 without explicit CPU specified are put on the local CPU. This 1396 guarantee is no longer true and while local CPU is still 1397 preferred work items may be put on foreign CPUs. Kernel 1398 parameter "workqueue.debug_force_rr_cpu" is added to force 1399 round-robin CPU selection to flush out usages which depend on the 1400 now broken guarantee. This config option enables the debug 1401 feature by default. When enabled, memory and cache locality will 1402 be impacted. 1403 1404config DEBUG_BLOCK_EXT_DEVT 1405 bool "Force extended block device numbers and spread them" 1406 depends on DEBUG_KERNEL 1407 depends on BLOCK 1408 default n 1409 help 1410 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1411 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1412 YOU ARE DOING. Distros, please enable this and fix whatever 1413 is broken. 1414 1415 Conventionally, block device numbers are allocated from 1416 predetermined contiguous area. However, extended block area 1417 may introduce non-contiguous block device numbers. This 1418 option forces most block device numbers to be allocated from 1419 the extended space and spreads them to discover kernel or 1420 userland code paths which assume predetermined contiguous 1421 device number allocation. 1422 1423 Note that turning on this debug option shuffles all the 1424 device numbers for all IDE and SCSI devices including libata 1425 ones, so root partition specified using device number 1426 directly (via rdev or root=MAJ:MIN) won't work anymore. 1427 Textual device names (root=/dev/sdXn) will continue to work. 1428 1429 Say N if you are unsure. 1430 1431config CPU_HOTPLUG_STATE_CONTROL 1432 bool "Enable CPU hotplug state control" 1433 depends on DEBUG_KERNEL 1434 depends on HOTPLUG_CPU 1435 default n 1436 help 1437 Allows to write steps between "offline" and "online" to the CPUs 1438 sysfs target file so states can be stepped granular. This is a debug 1439 option for now as the hotplug machinery cannot be stopped and 1440 restarted at arbitrary points yet. 1441 1442 Say N if your are unsure. 1443 1444config NOTIFIER_ERROR_INJECTION 1445 tristate "Notifier error injection" 1446 depends on DEBUG_KERNEL 1447 select DEBUG_FS 1448 help 1449 This option provides the ability to inject artificial errors to 1450 specified notifier chain callbacks. It is useful to test the error 1451 handling of notifier call chain failures. 1452 1453 Say N if unsure. 1454 1455config PM_NOTIFIER_ERROR_INJECT 1456 tristate "PM notifier error injection module" 1457 depends on PM && NOTIFIER_ERROR_INJECTION 1458 default m if PM_DEBUG 1459 help 1460 This option provides the ability to inject artificial errors to 1461 PM notifier chain callbacks. It is controlled through debugfs 1462 interface /sys/kernel/debug/notifier-error-inject/pm 1463 1464 If the notifier call chain should be failed with some events 1465 notified, write the error code to "actions/<notifier event>/error". 1466 1467 Example: Inject PM suspend error (-12 = -ENOMEM) 1468 1469 # cd /sys/kernel/debug/notifier-error-inject/pm/ 1470 # echo -12 > actions/PM_SUSPEND_PREPARE/error 1471 # echo mem > /sys/power/state 1472 bash: echo: write error: Cannot allocate memory 1473 1474 To compile this code as a module, choose M here: the module will 1475 be called pm-notifier-error-inject. 1476 1477 If unsure, say N. 1478 1479config OF_RECONFIG_NOTIFIER_ERROR_INJECT 1480 tristate "OF reconfig notifier error injection module" 1481 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 1482 help 1483 This option provides the ability to inject artificial errors to 1484 OF reconfig notifier chain callbacks. It is controlled 1485 through debugfs interface under 1486 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 1487 1488 If the notifier call chain should be failed with some events 1489 notified, write the error code to "actions/<notifier event>/error". 1490 1491 To compile this code as a module, choose M here: the module will 1492 be called of-reconfig-notifier-error-inject. 1493 1494 If unsure, say N. 1495 1496config NETDEV_NOTIFIER_ERROR_INJECT 1497 tristate "Netdev notifier error injection module" 1498 depends on NET && NOTIFIER_ERROR_INJECTION 1499 help 1500 This option provides the ability to inject artificial errors to 1501 netdevice notifier chain callbacks. It is controlled through debugfs 1502 interface /sys/kernel/debug/notifier-error-inject/netdev 1503 1504 If the notifier call chain should be failed with some events 1505 notified, write the error code to "actions/<notifier event>/error". 1506 1507 Example: Inject netdevice mtu change error (-22 = -EINVAL) 1508 1509 # cd /sys/kernel/debug/notifier-error-inject/netdev 1510 # echo -22 > actions/NETDEV_CHANGEMTU/error 1511 # ip link set eth0 mtu 1024 1512 RTNETLINK answers: Invalid argument 1513 1514 To compile this code as a module, choose M here: the module will 1515 be called netdev-notifier-error-inject. 1516 1517 If unsure, say N. 1518 1519config FAULT_INJECTION 1520 bool "Fault-injection framework" 1521 depends on DEBUG_KERNEL 1522 help 1523 Provide fault-injection framework. 1524 For more details, see Documentation/fault-injection/. 1525 1526config FAILSLAB 1527 bool "Fault-injection capability for kmalloc" 1528 depends on FAULT_INJECTION 1529 depends on SLAB || SLUB 1530 help 1531 Provide fault-injection capability for kmalloc. 1532 1533config FAIL_PAGE_ALLOC 1534 bool "Fault-injection capabilitiy for alloc_pages()" 1535 depends on FAULT_INJECTION 1536 help 1537 Provide fault-injection capability for alloc_pages(). 1538 1539config FAIL_MAKE_REQUEST 1540 bool "Fault-injection capability for disk IO" 1541 depends on FAULT_INJECTION && BLOCK 1542 help 1543 Provide fault-injection capability for disk IO. 1544 1545config FAIL_IO_TIMEOUT 1546 bool "Fault-injection capability for faking disk interrupts" 1547 depends on FAULT_INJECTION && BLOCK 1548 help 1549 Provide fault-injection capability on end IO handling. This 1550 will make the block layer "forget" an interrupt as configured, 1551 thus exercising the error handling. 1552 1553 Only works with drivers that use the generic timeout handling, 1554 for others it wont do anything. 1555 1556config FAIL_MMC_REQUEST 1557 bool "Fault-injection capability for MMC IO" 1558 depends on FAULT_INJECTION_DEBUG_FS && MMC 1559 help 1560 Provide fault-injection capability for MMC IO. 1561 This will make the mmc core return data errors. This is 1562 useful to test the error handling in the mmc block device 1563 and to test how the mmc host driver handles retries from 1564 the block device. 1565 1566config FAIL_FUTEX 1567 bool "Fault-injection capability for futexes" 1568 select DEBUG_FS 1569 depends on FAULT_INJECTION && FUTEX 1570 help 1571 Provide fault-injection capability for futexes. 1572 1573config FAULT_INJECTION_DEBUG_FS 1574 bool "Debugfs entries for fault-injection capabilities" 1575 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1576 help 1577 Enable configuration of fault-injection capabilities via debugfs. 1578 1579config FAULT_INJECTION_STACKTRACE_FILTER 1580 bool "stacktrace filter for fault-injection capabilities" 1581 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1582 depends on !X86_64 1583 select STACKTRACE 1584 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !SCORE && !X86 1585 help 1586 Provide stacktrace filter for fault-injection capabilities 1587 1588config LATENCYTOP 1589 bool "Latency measuring infrastructure" 1590 depends on DEBUG_KERNEL 1591 depends on STACKTRACE_SUPPORT 1592 depends on PROC_FS 1593 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !X86 1594 select KALLSYMS 1595 select KALLSYMS_ALL 1596 select STACKTRACE 1597 select SCHEDSTATS 1598 select SCHED_DEBUG 1599 help 1600 Enable this option if you want to use the LatencyTOP tool 1601 to find out which userspace is blocking on what kernel operations. 1602 1603source kernel/trace/Kconfig 1604 1605config PROVIDE_OHCI1394_DMA_INIT 1606 bool "Remote debugging over FireWire early on boot" 1607 depends on PCI && X86 1608 help 1609 If you want to debug problems which hang or crash the kernel early 1610 on boot and the crashing machine has a FireWire port, you can use 1611 this feature to remotely access the memory of the crashed machine 1612 over FireWire. This employs remote DMA as part of the OHCI1394 1613 specification which is now the standard for FireWire controllers. 1614 1615 With remote DMA, you can monitor the printk buffer remotely using 1616 firescope and access all memory below 4GB using fireproxy from gdb. 1617 Even controlling a kernel debugger is possible using remote DMA. 1618 1619 Usage: 1620 1621 If ohci1394_dma=early is used as boot parameter, it will initialize 1622 all OHCI1394 controllers which are found in the PCI config space. 1623 1624 As all changes to the FireWire bus such as enabling and disabling 1625 devices cause a bus reset and thereby disable remote DMA for all 1626 devices, be sure to have the cable plugged and FireWire enabled on 1627 the debugging host before booting the debug target for debugging. 1628 1629 This code (~1k) is freed after boot. By then, the firewire stack 1630 in charge of the OHCI-1394 controllers should be used instead. 1631 1632 See Documentation/debugging-via-ohci1394.txt for more information. 1633 1634config DMA_API_DEBUG 1635 bool "Enable debugging of DMA-API usage" 1636 depends on HAVE_DMA_API_DEBUG 1637 help 1638 Enable this option to debug the use of the DMA API by device drivers. 1639 With this option you will be able to detect common bugs in device 1640 drivers like double-freeing of DMA mappings or freeing mappings that 1641 were never allocated. 1642 1643 This also attempts to catch cases where a page owned by DMA is 1644 accessed by the cpu in a way that could cause data corruption. For 1645 example, this enables cow_user_page() to check that the source page is 1646 not undergoing DMA. 1647 1648 This option causes a performance degradation. Use only if you want to 1649 debug device drivers and dma interactions. 1650 1651 If unsure, say N. 1652 1653menu "Runtime Testing" 1654 1655config LKDTM 1656 tristate "Linux Kernel Dump Test Tool Module" 1657 depends on DEBUG_FS 1658 depends on BLOCK 1659 default n 1660 help 1661 This module enables testing of the different dumping mechanisms by 1662 inducing system failures at predefined crash points. 1663 If you don't need it: say N 1664 Choose M here to compile this code as a module. The module will be 1665 called lkdtm. 1666 1667 Documentation on how to use the module can be found in 1668 Documentation/fault-injection/provoke-crashes.txt 1669 1670config TEST_LIST_SORT 1671 tristate "Linked list sorting test" 1672 depends on DEBUG_KERNEL || m 1673 help 1674 Enable this to turn on 'list_sort()' function test. This test is 1675 executed only once during system boot (so affects only boot time), 1676 or at module load time. 1677 1678 If unsure, say N. 1679 1680config TEST_SORT 1681 tristate "Array-based sort test" 1682 depends on DEBUG_KERNEL || m 1683 help 1684 This option enables the self-test function of 'sort()' at boot, 1685 or at module load time. 1686 1687 If unsure, say N. 1688 1689config KPROBES_SANITY_TEST 1690 bool "Kprobes sanity tests" 1691 depends on DEBUG_KERNEL 1692 depends on KPROBES 1693 default n 1694 help 1695 This option provides for testing basic kprobes functionality on 1696 boot. A sample kprobe, jprobe and kretprobe are inserted and 1697 verified for functionality. 1698 1699 Say N if you are unsure. 1700 1701config BACKTRACE_SELF_TEST 1702 tristate "Self test for the backtrace code" 1703 depends on DEBUG_KERNEL 1704 default n 1705 help 1706 This option provides a kernel module that can be used to test 1707 the kernel stack backtrace code. This option is not useful 1708 for distributions or general kernels, but only for kernel 1709 developers working on architecture code. 1710 1711 Note that if you want to also test saved backtraces, you will 1712 have to enable STACKTRACE as well. 1713 1714 Say N if you are unsure. 1715 1716config RBTREE_TEST 1717 tristate "Red-Black tree test" 1718 depends on DEBUG_KERNEL 1719 help 1720 A benchmark measuring the performance of the rbtree library. 1721 Also includes rbtree invariant checks. 1722 1723config INTERVAL_TREE_TEST 1724 tristate "Interval tree test" 1725 depends on DEBUG_KERNEL 1726 select INTERVAL_TREE 1727 help 1728 A benchmark measuring the performance of the interval tree library 1729 1730config PERCPU_TEST 1731 tristate "Per cpu operations test" 1732 depends on m && DEBUG_KERNEL 1733 help 1734 Enable this option to build test module which validates per-cpu 1735 operations. 1736 1737 If unsure, say N. 1738 1739config ATOMIC64_SELFTEST 1740 tristate "Perform an atomic64_t self-test" 1741 help 1742 Enable this option to test the atomic64_t functions at boot or 1743 at module load time. 1744 1745 If unsure, say N. 1746 1747config ASYNC_RAID6_TEST 1748 tristate "Self test for hardware accelerated raid6 recovery" 1749 depends on ASYNC_RAID6_RECOV 1750 select ASYNC_MEMCPY 1751 ---help--- 1752 This is a one-shot self test that permutes through the 1753 recovery of all the possible two disk failure scenarios for a 1754 N-disk array. Recovery is performed with the asynchronous 1755 raid6 recovery routines, and will optionally use an offload 1756 engine if one is available. 1757 1758 If unsure, say N. 1759 1760config TEST_HEXDUMP 1761 tristate "Test functions located in the hexdump module at runtime" 1762 1763config TEST_STRING_HELPERS 1764 tristate "Test functions located in the string_helpers module at runtime" 1765 1766config TEST_KSTRTOX 1767 tristate "Test kstrto*() family of functions at runtime" 1768 1769config TEST_PRINTF 1770 tristate "Test printf() family of functions at runtime" 1771 1772config TEST_BITMAP 1773 tristate "Test bitmap_*() family of functions at runtime" 1774 default n 1775 help 1776 Enable this option to test the bitmap functions at boot. 1777 1778 If unsure, say N. 1779 1780config TEST_UUID 1781 tristate "Test functions located in the uuid module at runtime" 1782 1783config TEST_RHASHTABLE 1784 tristate "Perform selftest on resizable hash table" 1785 default n 1786 help 1787 Enable this option to test the rhashtable functions at boot. 1788 1789 If unsure, say N. 1790 1791config TEST_HASH 1792 tristate "Perform selftest on hash functions" 1793 default n 1794 help 1795 Enable this option to test the kernel's integer (<linux/hash.h>), 1796 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>) 1797 hash functions on boot (or module load). 1798 1799 This is intended to help people writing architecture-specific 1800 optimized versions. If unsure, say N. 1801 1802config TEST_PARMAN 1803 tristate "Perform selftest on priority array manager" 1804 default n 1805 depends on PARMAN 1806 help 1807 Enable this option to test priority array manager on boot 1808 (or module load). 1809 1810 If unsure, say N. 1811 1812config TEST_LKM 1813 tristate "Test module loading with 'hello world' module" 1814 default n 1815 depends on m 1816 help 1817 This builds the "test_module" module that emits "Hello, world" 1818 on printk when loaded. It is designed to be used for basic 1819 evaluation of the module loading subsystem (for example when 1820 validating module verification). It lacks any extra dependencies, 1821 and will not normally be loaded by the system unless explicitly 1822 requested by name. 1823 1824 If unsure, say N. 1825 1826config TEST_USER_COPY 1827 tristate "Test user/kernel boundary protections" 1828 default n 1829 depends on m 1830 help 1831 This builds the "test_user_copy" module that runs sanity checks 1832 on the copy_to/from_user infrastructure, making sure basic 1833 user/kernel boundary testing is working. If it fails to load, 1834 a regression has been detected in the user/kernel memory boundary 1835 protections. 1836 1837 If unsure, say N. 1838 1839config TEST_BPF 1840 tristate "Test BPF filter functionality" 1841 default n 1842 depends on m && NET 1843 help 1844 This builds the "test_bpf" module that runs various test vectors 1845 against the BPF interpreter or BPF JIT compiler depending on the 1846 current setting. This is in particular useful for BPF JIT compiler 1847 development, but also to run regression tests against changes in 1848 the interpreter code. It also enables test stubs for eBPF maps and 1849 verifier used by user space verifier testsuite. 1850 1851 If unsure, say N. 1852 1853config TEST_FIRMWARE 1854 tristate "Test firmware loading via userspace interface" 1855 default n 1856 depends on FW_LOADER 1857 help 1858 This builds the "test_firmware" module that creates a userspace 1859 interface for testing firmware loading. This can be used to 1860 control the triggering of firmware loading without needing an 1861 actual firmware-using device. The contents can be rechecked by 1862 userspace. 1863 1864 If unsure, say N. 1865 1866config TEST_SYSCTL 1867 tristate "sysctl test driver" 1868 default n 1869 depends on PROC_SYSCTL 1870 help 1871 This builds the "test_sysctl" module. This driver enables to test the 1872 proc sysctl interfaces available to drivers safely without affecting 1873 production knobs which might alter system functionality. 1874 1875 If unsure, say N. 1876 1877config TEST_UDELAY 1878 tristate "udelay test driver" 1879 default n 1880 help 1881 This builds the "udelay_test" module that helps to make sure 1882 that udelay() is working properly. 1883 1884 If unsure, say N. 1885 1886config TEST_STATIC_KEYS 1887 tristate "Test static keys" 1888 default n 1889 depends on m 1890 help 1891 Test the static key interfaces. 1892 1893 If unsure, say N. 1894 1895config TEST_KMOD 1896 tristate "kmod stress tester" 1897 default n 1898 depends on m 1899 depends on BLOCK && (64BIT || LBDAF) # for XFS, BTRFS 1900 depends on NETDEVICES && NET_CORE && INET # for TUN 1901 select TEST_LKM 1902 select XFS_FS 1903 select TUN 1904 select BTRFS_FS 1905 help 1906 Test the kernel's module loading mechanism: kmod. kmod implements 1907 support to load modules using the Linux kernel's usermode helper. 1908 This test provides a series of tests against kmod. 1909 1910 Although technically you can either build test_kmod as a module or 1911 into the kernel we disallow building it into the kernel since 1912 it stress tests request_module() and this will very likely cause 1913 some issues by taking over precious threads available from other 1914 module load requests, ultimately this could be fatal. 1915 1916 To run tests run: 1917 1918 tools/testing/selftests/kmod/kmod.sh --help 1919 1920 If unsure, say N. 1921 1922config TEST_DEBUG_VIRTUAL 1923 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 1924 depends on DEBUG_VIRTUAL 1925 help 1926 Test the kernel's ability to detect incorrect calls to 1927 virt_to_phys() done against the non-linear part of the 1928 kernel's virtual address map. 1929 1930 If unsure, say N. 1931 1932endmenu # runtime tests 1933 1934config MEMTEST 1935 bool "Memtest" 1936 depends on HAVE_MEMBLOCK 1937 ---help--- 1938 This option adds a kernel parameter 'memtest', which allows memtest 1939 to be set. 1940 memtest=0, mean disabled; -- default 1941 memtest=1, mean do 1 test pattern; 1942 ... 1943 memtest=17, mean do 17 test patterns. 1944 If you are unsure how to answer this question, answer N. 1945 1946config BUG_ON_DATA_CORRUPTION 1947 bool "Trigger a BUG when data corruption is detected" 1948 select DEBUG_LIST 1949 help 1950 Select this option if the kernel should BUG when it encounters 1951 data corruption in kernel memory structures when they get checked 1952 for validity. 1953 1954 If unsure, say N. 1955 1956source "samples/Kconfig" 1957 1958source "lib/Kconfig.kgdb" 1959 1960source "lib/Kconfig.ubsan" 1961 1962config ARCH_HAS_DEVMEM_IS_ALLOWED 1963 bool 1964 1965config STRICT_DEVMEM 1966 bool "Filter access to /dev/mem" 1967 depends on MMU && DEVMEM 1968 depends on ARCH_HAS_DEVMEM_IS_ALLOWED 1969 default y if TILE || PPC 1970 ---help--- 1971 If this option is disabled, you allow userspace (root) access to all 1972 of memory, including kernel and userspace memory. Accidental 1973 access to this is obviously disastrous, but specific access can 1974 be used by people debugging the kernel. Note that with PAT support 1975 enabled, even in this case there are restrictions on /dev/mem 1976 use due to the cache aliasing requirements. 1977 1978 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem 1979 file only allows userspace access to PCI space and the BIOS code and 1980 data regions. This is sufficient for dosemu and X and all common 1981 users of /dev/mem. 1982 1983 If in doubt, say Y. 1984 1985config IO_STRICT_DEVMEM 1986 bool "Filter I/O access to /dev/mem" 1987 depends on STRICT_DEVMEM 1988 ---help--- 1989 If this option is disabled, you allow userspace (root) access to all 1990 io-memory regardless of whether a driver is actively using that 1991 range. Accidental access to this is obviously disastrous, but 1992 specific access can be used by people debugging kernel drivers. 1993 1994 If this option is switched on, the /dev/mem file only allows 1995 userspace access to *idle* io-memory ranges (see /proc/iomem) This 1996 may break traditional users of /dev/mem (dosemu, legacy X, etc...) 1997 if the driver using a given range cannot be disabled. 1998 1999 If in doubt, say Y. 2000