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 select SRCU 284 help 285 debugfs is a virtual file system that kernel developers use to put 286 debugging files into. Enable this option to be able to read and 287 write to these files. 288 289 For detailed documentation on the debugfs API, see 290 Documentation/filesystems/. 291 292 If unsure, say N. 293 294config HEADERS_CHECK 295 bool "Run 'make headers_check' when building vmlinux" 296 depends on !UML 297 help 298 This option will extract the user-visible kernel headers whenever 299 building the kernel, and will run basic sanity checks on them to 300 ensure that exported files do not attempt to include files which 301 were not exported, etc. 302 303 If you're making modifications to header files which are 304 relevant for userspace, say 'Y', and check the headers 305 exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in 306 your build tree), to make sure they're suitable. 307 308config DEBUG_SECTION_MISMATCH 309 bool "Enable full Section mismatch analysis" 310 help 311 The section mismatch analysis checks if there are illegal 312 references from one section to another section. 313 During linktime or runtime, some sections are dropped; 314 any use of code/data previously in these sections would 315 most likely result in an oops. 316 In the code, functions and variables are annotated with 317 __init,, etc. (see the full list in include/linux/init.h), 318 which results in the code/data being placed in specific sections. 319 The section mismatch analysis is always performed after a full 320 kernel build, and enabling this option causes the following 321 additional steps to occur: 322 - Add the option -fno-inline-functions-called-once to gcc commands. 323 When inlining a function annotated with __init in a non-init 324 function, we would lose the section information and thus 325 the analysis would not catch the illegal reference. 326 This option tells gcc to inline less (but it does result in 327 a larger kernel). 328 - Run the section mismatch analysis for each module/built-in.o file. 329 When we run the section mismatch analysis on vmlinux.o, we 330 lose valuable information about where the mismatch was 331 introduced. 332 Running the analysis for each module/built-in.o file 333 tells where the mismatch happens much closer to the 334 source. The drawback is that the same mismatch is 335 reported at least twice. 336 - Enable verbose reporting from modpost in order to help resolve 337 the section mismatches that are reported. 338 339config SECTION_MISMATCH_WARN_ONLY 340 bool "Make section mismatch errors non-fatal" 341 default y 342 help 343 If you say N here, the build process will fail if there are any 344 section mismatch, instead of just throwing warnings. 345 346 If unsure, say Y. 347 348# 349# Select this config option from the architecture Kconfig, if it 350# is preferred to always offer frame pointers as a config 351# option on the architecture (regardless of KERNEL_DEBUG): 352# 353config ARCH_WANT_FRAME_POINTERS 354 bool 355 help 356 357config FRAME_POINTER 358 bool "Compile the kernel with frame pointers" 359 depends on DEBUG_KERNEL && \ 360 (CRIS || M68K || FRV || UML || \ 361 SUPERH || BLACKFIN || MN10300 || METAG) || \ 362 ARCH_WANT_FRAME_POINTERS 363 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 364 help 365 If you say Y here the resulting kernel image will be slightly 366 larger and slower, but it gives very useful debugging information 367 in case of kernel bugs. (precise oopses/stacktraces/warnings) 368 369config STACK_VALIDATION 370 bool "Compile-time stack metadata validation" 371 depends on HAVE_STACK_VALIDATION 372 default n 373 help 374 Add compile-time checks to validate stack metadata, including frame 375 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure 376 that runtime stack traces are more reliable. 377 378 This is also a prerequisite for generation of ORC unwind data, which 379 is needed for CONFIG_UNWINDER_ORC. 380 381 For more information, see 382 tools/objtool/Documentation/stack-validation.txt. 383 384config DEBUG_FORCE_WEAK_PER_CPU 385 bool "Force weak per-cpu definitions" 386 depends on DEBUG_KERNEL 387 help 388 s390 and alpha require percpu variables in modules to be 389 defined weak to work around addressing range issue which 390 puts the following two restrictions on percpu variable 391 definitions. 392 393 1. percpu symbols must be unique whether static or not 394 2. percpu variables can't be defined inside a function 395 396 To ensure that generic code follows the above rules, this 397 option forces all percpu variables to be defined as weak. 398 399endmenu # "Compiler options" 400 401config MAGIC_SYSRQ 402 bool "Magic SysRq key" 403 depends on !UML 404 help 405 If you say Y here, you will have some control over the system even 406 if the system crashes for example during kernel debugging (e.g., you 407 will be able to flush the buffer cache to disk, reboot the system 408 immediately or dump some status information). This is accomplished 409 by pressing various keys while holding SysRq (Alt+PrintScreen). It 410 also works on a serial console (on PC hardware at least), if you 411 send a BREAK and then within 5 seconds a command keypress. The 412 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 413 Don't say Y unless you really know what this hack does. 414 415config MAGIC_SYSRQ_DEFAULT_ENABLE 416 hex "Enable magic SysRq key functions by default" 417 depends on MAGIC_SYSRQ 418 default 0x1 419 help 420 Specifies which SysRq key functions are enabled by default. 421 This may be set to 1 or 0 to enable or disable them all, or 422 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 423 424config MAGIC_SYSRQ_SERIAL 425 bool "Enable magic SysRq key over serial" 426 depends on MAGIC_SYSRQ 427 default y 428 help 429 Many embedded boards have a disconnected TTL level serial which can 430 generate some garbage that can lead to spurious false sysrq detects. 431 This option allows you to decide whether you want to enable the 432 magic SysRq key. 433 434config DEBUG_KERNEL 435 bool "Kernel debugging" 436 help 437 Say Y here if you are developing drivers or trying to debug and 438 identify kernel problems. 439 440menu "Memory Debugging" 441 442source mm/Kconfig.debug 443 444config DEBUG_OBJECTS 445 bool "Debug object operations" 446 depends on DEBUG_KERNEL 447 help 448 If you say Y here, additional code will be inserted into the 449 kernel to track the life time of various objects and validate 450 the operations on those objects. 451 452config DEBUG_OBJECTS_SELFTEST 453 bool "Debug objects selftest" 454 depends on DEBUG_OBJECTS 455 help 456 This enables the selftest of the object debug code. 457 458config DEBUG_OBJECTS_FREE 459 bool "Debug objects in freed memory" 460 depends on DEBUG_OBJECTS 461 help 462 This enables checks whether a k/v free operation frees an area 463 which contains an object which has not been deactivated 464 properly. This can make kmalloc/kfree-intensive workloads 465 much slower. 466 467config DEBUG_OBJECTS_TIMERS 468 bool "Debug timer objects" 469 depends on DEBUG_OBJECTS 470 help 471 If you say Y here, additional code will be inserted into the 472 timer routines to track the life time of timer objects and 473 validate the timer operations. 474 475config DEBUG_OBJECTS_WORK 476 bool "Debug work objects" 477 depends on DEBUG_OBJECTS 478 help 479 If you say Y here, additional code will be inserted into the 480 work queue routines to track the life time of work objects and 481 validate the work operations. 482 483config DEBUG_OBJECTS_RCU_HEAD 484 bool "Debug RCU callbacks objects" 485 depends on DEBUG_OBJECTS 486 help 487 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 488 489config DEBUG_OBJECTS_PERCPU_COUNTER 490 bool "Debug percpu counter objects" 491 depends on DEBUG_OBJECTS 492 help 493 If you say Y here, additional code will be inserted into the 494 percpu counter routines to track the life time of percpu counter 495 objects and validate the percpu counter operations. 496 497config DEBUG_OBJECTS_ENABLE_DEFAULT 498 int "debug_objects bootup default value (0-1)" 499 range 0 1 500 default "1" 501 depends on DEBUG_OBJECTS 502 help 503 Debug objects boot parameter default value 504 505config DEBUG_SLAB 506 bool "Debug slab memory allocations" 507 depends on DEBUG_KERNEL && SLAB && !KMEMCHECK 508 help 509 Say Y here to have the kernel do limited verification on memory 510 allocation as well as poisoning memory on free to catch use of freed 511 memory. This can make kmalloc/kfree-intensive workloads much slower. 512 513config DEBUG_SLAB_LEAK 514 bool "Memory leak debugging" 515 depends on DEBUG_SLAB 516 517config SLUB_DEBUG_ON 518 bool "SLUB debugging on by default" 519 depends on SLUB && SLUB_DEBUG && !KMEMCHECK 520 default n 521 help 522 Boot with debugging on by default. SLUB boots by default with 523 the runtime debug capabilities switched off. Enabling this is 524 equivalent to specifying the "slub_debug" parameter on boot. 525 There is no support for more fine grained debug control like 526 possible with slub_debug=xxx. SLUB debugging may be switched 527 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 528 "slub_debug=-". 529 530config SLUB_STATS 531 default n 532 bool "Enable SLUB performance statistics" 533 depends on SLUB && SYSFS 534 help 535 SLUB statistics are useful to debug SLUBs allocation behavior in 536 order find ways to optimize the allocator. This should never be 537 enabled for production use since keeping statistics slows down 538 the allocator by a few percentage points. The slabinfo command 539 supports the determination of the most active slabs to figure 540 out which slabs are relevant to a particular load. 541 Try running: slabinfo -DA 542 543config HAVE_DEBUG_KMEMLEAK 544 bool 545 546config DEBUG_KMEMLEAK 547 bool "Kernel memory leak detector" 548 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 549 select DEBUG_FS 550 select STACKTRACE if STACKTRACE_SUPPORT 551 select KALLSYMS 552 select CRC32 553 help 554 Say Y here if you want to enable the memory leak 555 detector. The memory allocation/freeing is traced in a way 556 similar to the Boehm's conservative garbage collector, the 557 difference being that the orphan objects are not freed but 558 only shown in /sys/kernel/debug/kmemleak. Enabling this 559 feature will introduce an overhead to memory 560 allocations. See Documentation/dev-tools/kmemleak.rst for more 561 details. 562 563 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 564 of finding leaks due to the slab objects poisoning. 565 566 In order to access the kmemleak file, debugfs needs to be 567 mounted (usually at /sys/kernel/debug). 568 569config DEBUG_KMEMLEAK_EARLY_LOG_SIZE 570 int "Maximum kmemleak early log entries" 571 depends on DEBUG_KMEMLEAK 572 range 200 40000 573 default 400 574 help 575 Kmemleak must track all the memory allocations to avoid 576 reporting false positives. Since memory may be allocated or 577 freed before kmemleak is initialised, an early log buffer is 578 used to store these actions. If kmemleak reports "early log 579 buffer exceeded", please increase this value. 580 581config DEBUG_KMEMLEAK_TEST 582 tristate "Simple test for the kernel memory leak detector" 583 depends on DEBUG_KMEMLEAK && m 584 help 585 This option enables a module that explicitly leaks memory. 586 587 If unsure, say N. 588 589config DEBUG_KMEMLEAK_DEFAULT_OFF 590 bool "Default kmemleak to off" 591 depends on DEBUG_KMEMLEAK 592 help 593 Say Y here to disable kmemleak by default. It can then be enabled 594 on the command line via kmemleak=on. 595 596config DEBUG_STACK_USAGE 597 bool "Stack utilization instrumentation" 598 depends on DEBUG_KERNEL && !IA64 599 help 600 Enables the display of the minimum amount of free stack which each 601 task has ever had available in the sysrq-T and sysrq-P debug output. 602 603 This option will slow down process creation somewhat. 604 605config DEBUG_VM 606 bool "Debug VM" 607 depends on DEBUG_KERNEL 608 help 609 Enable this to turn on extended checks in the virtual-memory system 610 that may impact performance. 611 612 If unsure, say N. 613 614config DEBUG_VM_VMACACHE 615 bool "Debug VMA caching" 616 depends on DEBUG_VM 617 help 618 Enable this to turn on VMA caching debug information. Doing so 619 can cause significant overhead, so only enable it in non-production 620 environments. 621 622 If unsure, say N. 623 624config DEBUG_VM_RB 625 bool "Debug VM red-black trees" 626 depends on DEBUG_VM 627 help 628 Enable VM red-black tree debugging information and extra validations. 629 630 If unsure, say N. 631 632config DEBUG_VM_PGFLAGS 633 bool "Debug page-flags operations" 634 depends on DEBUG_VM 635 help 636 Enables extra validation on page flags operations. 637 638 If unsure, say N. 639 640config ARCH_HAS_DEBUG_VIRTUAL 641 bool 642 643config DEBUG_VIRTUAL 644 bool "Debug VM translations" 645 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 646 help 647 Enable some costly sanity checks in virtual to page code. This can 648 catch mistakes with virt_to_page() and friends. 649 650 If unsure, say N. 651 652config DEBUG_NOMMU_REGIONS 653 bool "Debug the global anon/private NOMMU mapping region tree" 654 depends on DEBUG_KERNEL && !MMU 655 help 656 This option causes the global tree of anonymous and private mapping 657 regions to be regularly checked for invalid topology. 658 659config DEBUG_MEMORY_INIT 660 bool "Debug memory initialisation" if EXPERT 661 default !EXPERT 662 help 663 Enable this for additional checks during memory initialisation. 664 The sanity checks verify aspects of the VM such as the memory model 665 and other information provided by the architecture. Verbose 666 information will be printed at KERN_DEBUG loglevel depending 667 on the mminit_loglevel= command-line option. 668 669 If unsure, say Y 670 671config MEMORY_NOTIFIER_ERROR_INJECT 672 tristate "Memory hotplug notifier error injection module" 673 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 674 help 675 This option provides the ability to inject artificial errors to 676 memory hotplug notifier chain callbacks. It is controlled through 677 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 678 679 If the notifier call chain should be failed with some events 680 notified, write the error code to "actions/<notifier event>/error". 681 682 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 683 684 # cd /sys/kernel/debug/notifier-error-inject/memory 685 # echo -12 > actions/MEM_GOING_OFFLINE/error 686 # echo offline > /sys/devices/system/memory/memoryXXX/state 687 bash: echo: write error: Cannot allocate memory 688 689 To compile this code as a module, choose M here: the module will 690 be called memory-notifier-error-inject. 691 692 If unsure, say N. 693 694config DEBUG_PER_CPU_MAPS 695 bool "Debug access to per_cpu maps" 696 depends on DEBUG_KERNEL 697 depends on SMP 698 help 699 Say Y to verify that the per_cpu map being accessed has 700 been set up. This adds a fair amount of code to kernel memory 701 and decreases performance. 702 703 Say N if unsure. 704 705config DEBUG_HIGHMEM 706 bool "Highmem debugging" 707 depends on DEBUG_KERNEL && HIGHMEM 708 help 709 This option enables additional error checking for high memory 710 systems. Disable for production systems. 711 712config HAVE_DEBUG_STACKOVERFLOW 713 bool 714 715config DEBUG_STACKOVERFLOW 716 bool "Check for stack overflows" 717 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 718 ---help--- 719 Say Y here if you want to check for overflows of kernel, IRQ 720 and exception stacks (if your architecture uses them). This 721 option will show detailed messages if free stack space drops 722 below a certain limit. 723 724 These kinds of bugs usually occur when call-chains in the 725 kernel get too deep, especially when interrupts are 726 involved. 727 728 Use this in cases where you see apparently random memory 729 corruption, especially if it appears in 'struct thread_info' 730 731 If in doubt, say "N". 732 733source "lib/Kconfig.kmemcheck" 734 735source "lib/Kconfig.kasan" 736 737endmenu # "Memory Debugging" 738 739config ARCH_HAS_KCOV 740 bool 741 help 742 KCOV does not have any arch-specific code, but currently it is enabled 743 only for x86_64. KCOV requires testing on other archs, and most likely 744 disabling of instrumentation for some early boot code. 745 746config KCOV 747 bool "Code coverage for fuzzing" 748 depends on ARCH_HAS_KCOV 749 select DEBUG_FS 750 select GCC_PLUGINS if !COMPILE_TEST 751 select GCC_PLUGIN_SANCOV if !COMPILE_TEST 752 help 753 KCOV exposes kernel code coverage information in a form suitable 754 for coverage-guided fuzzing (randomized testing). 755 756 If RANDOMIZE_BASE is enabled, PC values will not be stable across 757 different machines and across reboots. If you need stable PC values, 758 disable RANDOMIZE_BASE. 759 760 For more details, see Documentation/dev-tools/kcov.rst. 761 762config KCOV_INSTRUMENT_ALL 763 bool "Instrument all code by default" 764 depends on KCOV 765 default y if KCOV 766 help 767 If you are doing generic system call fuzzing (like e.g. syzkaller), 768 then you will want to instrument the whole kernel and you should 769 say y here. If you are doing more targeted fuzzing (like e.g. 770 filesystem fuzzing with AFL) then you will want to enable coverage 771 for more specific subsets of files, and should say n here. 772 773config DEBUG_SHIRQ 774 bool "Debug shared IRQ handlers" 775 depends on DEBUG_KERNEL 776 help 777 Enable this to generate a spurious interrupt as soon as a shared 778 interrupt handler is registered, and just before one is deregistered. 779 Drivers ought to be able to handle interrupts coming in at those 780 points; some don't and need to be caught. 781 782menu "Debug Lockups and Hangs" 783 784config LOCKUP_DETECTOR 785 bool 786 787config SOFTLOCKUP_DETECTOR 788 bool "Detect Soft Lockups" 789 depends on DEBUG_KERNEL && !S390 790 select LOCKUP_DETECTOR 791 help 792 Say Y here to enable the kernel to act as a watchdog to detect 793 soft lockups. 794 795 Softlockups are bugs that cause the kernel to loop in kernel 796 mode for more than 20 seconds, without giving other tasks a 797 chance to run. The current stack trace is displayed upon 798 detection and the system will stay locked up. 799 800config HARDLOCKUP_DETECTOR_PERF 801 bool 802 select SOFTLOCKUP_DETECTOR 803 804# 805# Enables a timestamp based low pass filter to compensate for perf based 806# hard lockup detection which runs too fast due to turbo modes. 807# 808config HARDLOCKUP_CHECK_TIMESTAMP 809 bool 810 811# 812# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard 813# lockup detector rather than the perf based detector. 814# 815config HARDLOCKUP_DETECTOR 816 bool "Detect Hard Lockups" 817 depends on DEBUG_KERNEL && !S390 818 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH 819 select LOCKUP_DETECTOR 820 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF 821 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH 822 help 823 Say Y here to enable the kernel to act as a watchdog to detect 824 hard lockups. 825 826 Hardlockups are bugs that cause the CPU to loop in kernel mode 827 for more than 10 seconds, without letting other interrupts have a 828 chance to run. The current stack trace is displayed upon detection 829 and the system will stay locked up. 830 831config BOOTPARAM_HARDLOCKUP_PANIC 832 bool "Panic (Reboot) On Hard Lockups" 833 depends on HARDLOCKUP_DETECTOR 834 help 835 Say Y here to enable the kernel to panic on "hard lockups", 836 which are bugs that cause the kernel to loop in kernel 837 mode with interrupts disabled for more than 10 seconds (configurable 838 using the watchdog_thresh sysctl). 839 840 Say N if unsure. 841 842config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 843 int 844 depends on HARDLOCKUP_DETECTOR 845 range 0 1 846 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 847 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 848 849config BOOTPARAM_SOFTLOCKUP_PANIC 850 bool "Panic (Reboot) On Soft Lockups" 851 depends on SOFTLOCKUP_DETECTOR 852 help 853 Say Y here to enable the kernel to panic on "soft lockups", 854 which are bugs that cause the kernel to loop in kernel 855 mode for more than 20 seconds (configurable using the watchdog_thresh 856 sysctl), without giving other tasks a chance to run. 857 858 The panic can be used in combination with panic_timeout, 859 to cause the system to reboot automatically after a 860 lockup has been detected. This feature is useful for 861 high-availability systems that have uptime guarantees and 862 where a lockup must be resolved ASAP. 863 864 Say N if unsure. 865 866config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 867 int 868 depends on SOFTLOCKUP_DETECTOR 869 range 0 1 870 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 871 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 872 873config DETECT_HUNG_TASK 874 bool "Detect Hung Tasks" 875 depends on DEBUG_KERNEL 876 default SOFTLOCKUP_DETECTOR 877 help 878 Say Y here to enable the kernel to detect "hung tasks", 879 which are bugs that cause the task to be stuck in 880 uninterruptible "D" state indefinitely. 881 882 When a hung task is detected, the kernel will print the 883 current stack trace (which you should report), but the 884 task will stay in uninterruptible state. If lockdep is 885 enabled then all held locks will also be reported. This 886 feature has negligible overhead. 887 888config DEFAULT_HUNG_TASK_TIMEOUT 889 int "Default timeout for hung task detection (in seconds)" 890 depends on DETECT_HUNG_TASK 891 default 120 892 help 893 This option controls the default timeout (in seconds) used 894 to determine when a task has become non-responsive and should 895 be considered hung. 896 897 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 898 sysctl or by writing a value to 899 /proc/sys/kernel/hung_task_timeout_secs. 900 901 A timeout of 0 disables the check. The default is two minutes. 902 Keeping the default should be fine in most cases. 903 904config BOOTPARAM_HUNG_TASK_PANIC 905 bool "Panic (Reboot) On Hung Tasks" 906 depends on DETECT_HUNG_TASK 907 help 908 Say Y here to enable the kernel to panic on "hung tasks", 909 which are bugs that cause the kernel to leave a task stuck 910 in uninterruptible "D" state. 911 912 The panic can be used in combination with panic_timeout, 913 to cause the system to reboot automatically after a 914 hung task has been detected. This feature is useful for 915 high-availability systems that have uptime guarantees and 916 where a hung tasks must be resolved ASAP. 917 918 Say N if unsure. 919 920config BOOTPARAM_HUNG_TASK_PANIC_VALUE 921 int 922 depends on DETECT_HUNG_TASK 923 range 0 1 924 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 925 default 1 if BOOTPARAM_HUNG_TASK_PANIC 926 927config WQ_WATCHDOG 928 bool "Detect Workqueue Stalls" 929 depends on DEBUG_KERNEL 930 help 931 Say Y here to enable stall detection on workqueues. If a 932 worker pool doesn't make forward progress on a pending work 933 item for over a given amount of time, 30s by default, a 934 warning message is printed along with dump of workqueue 935 state. This can be configured through kernel parameter 936 "workqueue.watchdog_thresh" and its sysfs counterpart. 937 938endmenu # "Debug lockups and hangs" 939 940config PANIC_ON_OOPS 941 bool "Panic on Oops" 942 help 943 Say Y here to enable the kernel to panic when it oopses. This 944 has the same effect as setting oops=panic on the kernel command 945 line. 946 947 This feature is useful to ensure that the kernel does not do 948 anything erroneous after an oops which could result in data 949 corruption or other issues. 950 951 Say N if unsure. 952 953config PANIC_ON_OOPS_VALUE 954 int 955 range 0 1 956 default 0 if !PANIC_ON_OOPS 957 default 1 if PANIC_ON_OOPS 958 959config PANIC_TIMEOUT 960 int "panic timeout" 961 default 0 962 help 963 Set the timeout value (in seconds) until a reboot occurs when the 964 the kernel panics. If n = 0, then we wait forever. A timeout 965 value n > 0 will wait n seconds before rebooting, while a timeout 966 value n < 0 will reboot immediately. 967 968config SCHED_DEBUG 969 bool "Collect scheduler debugging info" 970 depends on DEBUG_KERNEL && PROC_FS 971 default y 972 help 973 If you say Y here, the /proc/sched_debug file will be provided 974 that can help debug the scheduler. The runtime overhead of this 975 option is minimal. 976 977config SCHED_INFO 978 bool 979 default n 980 981config SCHEDSTATS 982 bool "Collect scheduler statistics" 983 depends on DEBUG_KERNEL && PROC_FS 984 select SCHED_INFO 985 help 986 If you say Y here, additional code will be inserted into the 987 scheduler and related routines to collect statistics about 988 scheduler behavior and provide them in /proc/schedstat. These 989 stats may be useful for both tuning and debugging the scheduler 990 If you aren't debugging the scheduler or trying to tune a specific 991 application, you can say N to avoid the very slight overhead 992 this adds. 993 994config SCHED_STACK_END_CHECK 995 bool "Detect stack corruption on calls to schedule()" 996 depends on DEBUG_KERNEL 997 default n 998 help 999 This option checks for a stack overrun on calls to schedule(). 1000 If the stack end location is found to be over written always panic as 1001 the content of the corrupted region can no longer be trusted. 1002 This is to ensure no erroneous behaviour occurs which could result in 1003 data corruption or a sporadic crash at a later stage once the region 1004 is examined. The runtime overhead introduced is minimal. 1005 1006config DEBUG_TIMEKEEPING 1007 bool "Enable extra timekeeping sanity checking" 1008 help 1009 This option will enable additional timekeeping sanity checks 1010 which may be helpful when diagnosing issues where timekeeping 1011 problems are suspected. 1012 1013 This may include checks in the timekeeping hotpaths, so this 1014 option may have a (very small) performance impact to some 1015 workloads. 1016 1017 If unsure, say N. 1018 1019config DEBUG_PREEMPT 1020 bool "Debug preemptible kernel" 1021 depends on DEBUG_KERNEL && PREEMPT && TRACE_IRQFLAGS_SUPPORT 1022 default y 1023 help 1024 If you say Y here then the kernel will use a debug variant of the 1025 commonly used smp_processor_id() function and will print warnings 1026 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1027 will detect preemption count underflows. 1028 1029menu "Lock Debugging (spinlocks, mutexes, etc...)" 1030 1031config DEBUG_RT_MUTEXES 1032 bool "RT Mutex debugging, deadlock detection" 1033 depends on DEBUG_KERNEL && RT_MUTEXES 1034 help 1035 This allows rt mutex semantics violations and rt mutex related 1036 deadlocks (lockups) to be detected and reported automatically. 1037 1038config DEBUG_SPINLOCK 1039 bool "Spinlock and rw-lock debugging: basic checks" 1040 depends on DEBUG_KERNEL 1041 select UNINLINE_SPIN_UNLOCK 1042 help 1043 Say Y here and build SMP to catch missing spinlock initialization 1044 and certain other kinds of spinlock errors commonly made. This is 1045 best used in conjunction with the NMI watchdog so that spinlock 1046 deadlocks are also debuggable. 1047 1048config DEBUG_MUTEXES 1049 bool "Mutex debugging: basic checks" 1050 depends on DEBUG_KERNEL 1051 help 1052 This feature allows mutex semantics violations to be detected and 1053 reported. 1054 1055config DEBUG_WW_MUTEX_SLOWPATH 1056 bool "Wait/wound mutex debugging: Slowpath testing" 1057 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1058 select DEBUG_LOCK_ALLOC 1059 select DEBUG_SPINLOCK 1060 select DEBUG_MUTEXES 1061 help 1062 This feature enables slowpath testing for w/w mutex users by 1063 injecting additional -EDEADLK wound/backoff cases. Together with 1064 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1065 will test all possible w/w mutex interface abuse with the 1066 exception of simply not acquiring all the required locks. 1067 Note that this feature can introduce significant overhead, so 1068 it really should not be enabled in a production or distro kernel, 1069 even a debug kernel. If you are a driver writer, enable it. If 1070 you are a distro, do not. 1071 1072config DEBUG_LOCK_ALLOC 1073 bool "Lock debugging: detect incorrect freeing of live locks" 1074 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1075 select DEBUG_SPINLOCK 1076 select DEBUG_MUTEXES 1077 select DEBUG_RT_MUTEXES if RT_MUTEXES 1078 select LOCKDEP 1079 help 1080 This feature will check whether any held lock (spinlock, rwlock, 1081 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1082 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1083 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1084 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1085 held during task exit. 1086 1087config PROVE_LOCKING 1088 bool "Lock debugging: prove locking correctness" 1089 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1090 select LOCKDEP 1091 select DEBUG_SPINLOCK 1092 select DEBUG_MUTEXES 1093 select DEBUG_RT_MUTEXES if RT_MUTEXES 1094 select DEBUG_LOCK_ALLOC 1095 select LOCKDEP_CROSSRELEASE 1096 select LOCKDEP_COMPLETIONS 1097 select TRACE_IRQFLAGS 1098 default n 1099 help 1100 This feature enables the kernel to prove that all locking 1101 that occurs in the kernel runtime is mathematically 1102 correct: that under no circumstance could an arbitrary (and 1103 not yet triggered) combination of observed locking 1104 sequences (on an arbitrary number of CPUs, running an 1105 arbitrary number of tasks and interrupt contexts) cause a 1106 deadlock. 1107 1108 In short, this feature enables the kernel to report locking 1109 related deadlocks before they actually occur. 1110 1111 The proof does not depend on how hard and complex a 1112 deadlock scenario would be to trigger: how many 1113 participant CPUs, tasks and irq-contexts would be needed 1114 for it to trigger. The proof also does not depend on 1115 timing: if a race and a resulting deadlock is possible 1116 theoretically (no matter how unlikely the race scenario 1117 is), it will be proven so and will immediately be 1118 reported by the kernel (once the event is observed that 1119 makes the deadlock theoretically possible). 1120 1121 If a deadlock is impossible (i.e. the locking rules, as 1122 observed by the kernel, are mathematically correct), the 1123 kernel reports nothing. 1124 1125 NOTE: this feature can also be enabled for rwlocks, mutexes 1126 and rwsems - in which case all dependencies between these 1127 different locking variants are observed and mapped too, and 1128 the proof of observed correctness is also maintained for an 1129 arbitrary combination of these separate locking variants. 1130 1131 For more details, see Documentation/locking/lockdep-design.txt. 1132 1133config LOCKDEP 1134 bool 1135 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1136 select STACKTRACE 1137 select FRAME_POINTER if !MIPS && !PPC && !ARM_UNWIND && !S390 && !MICROBLAZE && !ARC && !SCORE && !X86 1138 select KALLSYMS 1139 select KALLSYMS_ALL 1140 1141config LOCKDEP_SMALL 1142 bool 1143 1144config LOCK_STAT 1145 bool "Lock usage statistics" 1146 depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1147 select LOCKDEP 1148 select DEBUG_SPINLOCK 1149 select DEBUG_MUTEXES 1150 select DEBUG_RT_MUTEXES if RT_MUTEXES 1151 select DEBUG_LOCK_ALLOC 1152 default n 1153 help 1154 This feature enables tracking lock contention points 1155 1156 For more details, see Documentation/locking/lockstat.txt 1157 1158 This also enables lock events required by "perf lock", 1159 subcommand of perf. 1160 If you want to use "perf lock", you also need to turn on 1161 CONFIG_EVENT_TRACING. 1162 1163 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1164 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1165 1166config LOCKDEP_CROSSRELEASE 1167 bool 1168 help 1169 This makes lockdep work for crosslock which is a lock allowed to 1170 be released in a different context from the acquisition context. 1171 Normally a lock must be released in the context acquiring the lock. 1172 However, relexing this constraint helps synchronization primitives 1173 such as page locks or completions can use the lock correctness 1174 detector, lockdep. 1175 1176config LOCKDEP_COMPLETIONS 1177 bool 1178 help 1179 A deadlock caused by wait_for_completion() and complete() can be 1180 detected by lockdep using crossrelease feature. 1181 1182config BOOTPARAM_LOCKDEP_CROSSRELEASE_FULLSTACK 1183 bool "Enable the boot parameter, crossrelease_fullstack" 1184 depends on LOCKDEP_CROSSRELEASE 1185 default n 1186 help 1187 The lockdep "cross-release" feature needs to record stack traces 1188 (of calling functions) for all acquisitions, for eventual later 1189 use during analysis. By default only a single caller is recorded, 1190 because the unwind operation can be very expensive with deeper 1191 stack chains. 1192 1193 However a boot parameter, crossrelease_fullstack, was 1194 introduced since sometimes deeper traces are required for full 1195 analysis. This option turns on the boot parameter. 1196 1197config DEBUG_LOCKDEP 1198 bool "Lock dependency engine debugging" 1199 depends on DEBUG_KERNEL && LOCKDEP 1200 help 1201 If you say Y here, the lock dependency engine will do 1202 additional runtime checks to debug itself, at the price 1203 of more runtime overhead. 1204 1205config DEBUG_ATOMIC_SLEEP 1206 bool "Sleep inside atomic section checking" 1207 select PREEMPT_COUNT 1208 depends on DEBUG_KERNEL 1209 help 1210 If you say Y here, various routines which may sleep will become very 1211 noisy if they are called inside atomic sections: when a spinlock is 1212 held, inside an rcu read side critical section, inside preempt disabled 1213 sections, inside an interrupt, etc... 1214 1215config DEBUG_LOCKING_API_SELFTESTS 1216 bool "Locking API boot-time self-tests" 1217 depends on DEBUG_KERNEL 1218 help 1219 Say Y here if you want the kernel to run a short self-test during 1220 bootup. The self-test checks whether common types of locking bugs 1221 are detected by debugging mechanisms or not. (if you disable 1222 lock debugging then those bugs wont be detected of course.) 1223 The following locking APIs are covered: spinlocks, rwlocks, 1224 mutexes and rwsems. 1225 1226config LOCK_TORTURE_TEST 1227 tristate "torture tests for locking" 1228 depends on DEBUG_KERNEL 1229 select TORTURE_TEST 1230 default n 1231 help 1232 This option provides a kernel module that runs torture tests 1233 on kernel locking primitives. The kernel module may be built 1234 after the fact on the running kernel to be tested, if desired. 1235 1236 Say Y here if you want kernel locking-primitive torture tests 1237 to be built into the kernel. 1238 Say M if you want these torture tests to build as a module. 1239 Say N if you are unsure. 1240 1241config WW_MUTEX_SELFTEST 1242 tristate "Wait/wound mutex selftests" 1243 help 1244 This option provides a kernel module that runs tests on the 1245 on the struct ww_mutex locking API. 1246 1247 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1248 with this test harness. 1249 1250 Say M if you want these self tests to build as a module. 1251 Say N if you are unsure. 1252 1253endmenu # lock debugging 1254 1255config TRACE_IRQFLAGS 1256 bool 1257 help 1258 Enables hooks to interrupt enabling and disabling for 1259 either tracing or lock debugging. 1260 1261config STACKTRACE 1262 bool "Stack backtrace support" 1263 depends on STACKTRACE_SUPPORT 1264 help 1265 This option causes the kernel to create a /proc/pid/stack for 1266 every process, showing its current stack trace. 1267 It is also used by various kernel debugging features that require 1268 stack trace generation. 1269 1270config WARN_ALL_UNSEEDED_RANDOM 1271 bool "Warn for all uses of unseeded randomness" 1272 default n 1273 help 1274 Some parts of the kernel contain bugs relating to their use of 1275 cryptographically secure random numbers before it's actually possible 1276 to generate those numbers securely. This setting ensures that these 1277 flaws don't go unnoticed, by enabling a message, should this ever 1278 occur. This will allow people with obscure setups to know when things 1279 are going wrong, so that they might contact developers about fixing 1280 it. 1281 1282 Unfortunately, on some models of some architectures getting 1283 a fully seeded CRNG is extremely difficult, and so this can 1284 result in dmesg getting spammed for a surprisingly long 1285 time. This is really bad from a security perspective, and 1286 so architecture maintainers really need to do what they can 1287 to get the CRNG seeded sooner after the system is booted. 1288 However, since users can not do anything actionble to 1289 address this, by default the kernel will issue only a single 1290 warning for the first use of unseeded randomness. 1291 1292 Say Y here if you want to receive warnings for all uses of 1293 unseeded randomness. This will be of use primarily for 1294 those developers interersted in improving the security of 1295 Linux kernels running on their architecture (or 1296 subarchitecture). 1297 1298config DEBUG_KOBJECT 1299 bool "kobject debugging" 1300 depends on DEBUG_KERNEL 1301 help 1302 If you say Y here, some extra kobject debugging messages will be sent 1303 to the syslog. 1304 1305config DEBUG_KOBJECT_RELEASE 1306 bool "kobject release debugging" 1307 depends on DEBUG_OBJECTS_TIMERS 1308 help 1309 kobjects are reference counted objects. This means that their 1310 last reference count put is not predictable, and the kobject can 1311 live on past the point at which a driver decides to drop it's 1312 initial reference to the kobject gained on allocation. An 1313 example of this would be a struct device which has just been 1314 unregistered. 1315 1316 However, some buggy drivers assume that after such an operation, 1317 the memory backing the kobject can be immediately freed. This 1318 goes completely against the principles of a refcounted object. 1319 1320 If you say Y here, the kernel will delay the release of kobjects 1321 on the last reference count to improve the visibility of this 1322 kind of kobject release bug. 1323 1324config HAVE_DEBUG_BUGVERBOSE 1325 bool 1326 1327config DEBUG_BUGVERBOSE 1328 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 1329 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE) 1330 default y 1331 help 1332 Say Y here to make BUG() panics output the file name and line number 1333 of the BUG call as well as the EIP and oops trace. This aids 1334 debugging but costs about 70-100K of memory. 1335 1336config DEBUG_LIST 1337 bool "Debug linked list manipulation" 1338 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION 1339 help 1340 Enable this to turn on extended checks in the linked-list 1341 walking routines. 1342 1343 If unsure, say N. 1344 1345config DEBUG_PI_LIST 1346 bool "Debug priority linked list manipulation" 1347 depends on DEBUG_KERNEL 1348 help 1349 Enable this to turn on extended checks in the priority-ordered 1350 linked-list (plist) walking routines. This checks the entire 1351 list multiple times during each manipulation. 1352 1353 If unsure, say N. 1354 1355config DEBUG_SG 1356 bool "Debug SG table operations" 1357 depends on DEBUG_KERNEL 1358 help 1359 Enable this to turn on checks on scatter-gather tables. This can 1360 help find problems with drivers that do not properly initialize 1361 their sg tables. 1362 1363 If unsure, say N. 1364 1365config DEBUG_NOTIFIERS 1366 bool "Debug notifier call chains" 1367 depends on DEBUG_KERNEL 1368 help 1369 Enable this to turn on sanity checking for notifier call chains. 1370 This is most useful for kernel developers to make sure that 1371 modules properly unregister themselves from notifier chains. 1372 This is a relatively cheap check but if you care about maximum 1373 performance, say N. 1374 1375config DEBUG_CREDENTIALS 1376 bool "Debug credential management" 1377 depends on DEBUG_KERNEL 1378 help 1379 Enable this to turn on some debug checking for credential 1380 management. The additional code keeps track of the number of 1381 pointers from task_structs to any given cred struct, and checks to 1382 see that this number never exceeds the usage count of the cred 1383 struct. 1384 1385 Furthermore, if SELinux is enabled, this also checks that the 1386 security pointer in the cred struct is never seen to be invalid. 1387 1388 If unsure, say N. 1389 1390source "kernel/rcu/Kconfig.debug" 1391 1392config DEBUG_WQ_FORCE_RR_CPU 1393 bool "Force round-robin CPU selection for unbound work items" 1394 depends on DEBUG_KERNEL 1395 default n 1396 help 1397 Workqueue used to implicitly guarantee that work items queued 1398 without explicit CPU specified are put on the local CPU. This 1399 guarantee is no longer true and while local CPU is still 1400 preferred work items may be put on foreign CPUs. Kernel 1401 parameter "workqueue.debug_force_rr_cpu" is added to force 1402 round-robin CPU selection to flush out usages which depend on the 1403 now broken guarantee. This config option enables the debug 1404 feature by default. When enabled, memory and cache locality will 1405 be impacted. 1406 1407config DEBUG_BLOCK_EXT_DEVT 1408 bool "Force extended block device numbers and spread them" 1409 depends on DEBUG_KERNEL 1410 depends on BLOCK 1411 default n 1412 help 1413 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1414 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1415 YOU ARE DOING. Distros, please enable this and fix whatever 1416 is broken. 1417 1418 Conventionally, block device numbers are allocated from 1419 predetermined contiguous area. However, extended block area 1420 may introduce non-contiguous block device numbers. This 1421 option forces most block device numbers to be allocated from 1422 the extended space and spreads them to discover kernel or 1423 userland code paths which assume predetermined contiguous 1424 device number allocation. 1425 1426 Note that turning on this debug option shuffles all the 1427 device numbers for all IDE and SCSI devices including libata 1428 ones, so root partition specified using device number 1429 directly (via rdev or root=MAJ:MIN) won't work anymore. 1430 Textual device names (root=/dev/sdXn) will continue to work. 1431 1432 Say N if you are unsure. 1433 1434config CPU_HOTPLUG_STATE_CONTROL 1435 bool "Enable CPU hotplug state control" 1436 depends on DEBUG_KERNEL 1437 depends on HOTPLUG_CPU 1438 default n 1439 help 1440 Allows to write steps between "offline" and "online" to the CPUs 1441 sysfs target file so states can be stepped granular. This is a debug 1442 option for now as the hotplug machinery cannot be stopped and 1443 restarted at arbitrary points yet. 1444 1445 Say N if your are unsure. 1446 1447config NOTIFIER_ERROR_INJECTION 1448 tristate "Notifier error injection" 1449 depends on DEBUG_KERNEL 1450 select DEBUG_FS 1451 help 1452 This option provides the ability to inject artificial errors to 1453 specified notifier chain callbacks. It is useful to test the error 1454 handling of notifier call chain failures. 1455 1456 Say N if unsure. 1457 1458config PM_NOTIFIER_ERROR_INJECT 1459 tristate "PM notifier error injection module" 1460 depends on PM && NOTIFIER_ERROR_INJECTION 1461 default m if PM_DEBUG 1462 help 1463 This option provides the ability to inject artificial errors to 1464 PM notifier chain callbacks. It is controlled through debugfs 1465 interface /sys/kernel/debug/notifier-error-inject/pm 1466 1467 If the notifier call chain should be failed with some events 1468 notified, write the error code to "actions/<notifier event>/error". 1469 1470 Example: Inject PM suspend error (-12 = -ENOMEM) 1471 1472 # cd /sys/kernel/debug/notifier-error-inject/pm/ 1473 # echo -12 > actions/PM_SUSPEND_PREPARE/error 1474 # echo mem > /sys/power/state 1475 bash: echo: write error: Cannot allocate memory 1476 1477 To compile this code as a module, choose M here: the module will 1478 be called pm-notifier-error-inject. 1479 1480 If unsure, say N. 1481 1482config OF_RECONFIG_NOTIFIER_ERROR_INJECT 1483 tristate "OF reconfig notifier error injection module" 1484 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 1485 help 1486 This option provides the ability to inject artificial errors to 1487 OF reconfig notifier chain callbacks. It is controlled 1488 through debugfs interface under 1489 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 1490 1491 If the notifier call chain should be failed with some events 1492 notified, write the error code to "actions/<notifier event>/error". 1493 1494 To compile this code as a module, choose M here: the module will 1495 be called of-reconfig-notifier-error-inject. 1496 1497 If unsure, say N. 1498 1499config NETDEV_NOTIFIER_ERROR_INJECT 1500 tristate "Netdev notifier error injection module" 1501 depends on NET && NOTIFIER_ERROR_INJECTION 1502 help 1503 This option provides the ability to inject artificial errors to 1504 netdevice notifier chain callbacks. It is controlled through debugfs 1505 interface /sys/kernel/debug/notifier-error-inject/netdev 1506 1507 If the notifier call chain should be failed with some events 1508 notified, write the error code to "actions/<notifier event>/error". 1509 1510 Example: Inject netdevice mtu change error (-22 = -EINVAL) 1511 1512 # cd /sys/kernel/debug/notifier-error-inject/netdev 1513 # echo -22 > actions/NETDEV_CHANGEMTU/error 1514 # ip link set eth0 mtu 1024 1515 RTNETLINK answers: Invalid argument 1516 1517 To compile this code as a module, choose M here: the module will 1518 be called netdev-notifier-error-inject. 1519 1520 If unsure, say N. 1521 1522config FAULT_INJECTION 1523 bool "Fault-injection framework" 1524 depends on DEBUG_KERNEL 1525 help 1526 Provide fault-injection framework. 1527 For more details, see Documentation/fault-injection/. 1528 1529config FAILSLAB 1530 bool "Fault-injection capability for kmalloc" 1531 depends on FAULT_INJECTION 1532 depends on SLAB || SLUB 1533 help 1534 Provide fault-injection capability for kmalloc. 1535 1536config FAIL_PAGE_ALLOC 1537 bool "Fault-injection capabilitiy for alloc_pages()" 1538 depends on FAULT_INJECTION 1539 help 1540 Provide fault-injection capability for alloc_pages(). 1541 1542config FAIL_MAKE_REQUEST 1543 bool "Fault-injection capability for disk IO" 1544 depends on FAULT_INJECTION && BLOCK 1545 help 1546 Provide fault-injection capability for disk IO. 1547 1548config FAIL_IO_TIMEOUT 1549 bool "Fault-injection capability for faking disk interrupts" 1550 depends on FAULT_INJECTION && BLOCK 1551 help 1552 Provide fault-injection capability on end IO handling. This 1553 will make the block layer "forget" an interrupt as configured, 1554 thus exercising the error handling. 1555 1556 Only works with drivers that use the generic timeout handling, 1557 for others it wont do anything. 1558 1559config FAIL_MMC_REQUEST 1560 bool "Fault-injection capability for MMC IO" 1561 depends on FAULT_INJECTION_DEBUG_FS && MMC 1562 help 1563 Provide fault-injection capability for MMC IO. 1564 This will make the mmc core return data errors. This is 1565 useful to test the error handling in the mmc block device 1566 and to test how the mmc host driver handles retries from 1567 the block device. 1568 1569config FAIL_FUTEX 1570 bool "Fault-injection capability for futexes" 1571 select DEBUG_FS 1572 depends on FAULT_INJECTION && FUTEX 1573 help 1574 Provide fault-injection capability for futexes. 1575 1576config FAULT_INJECTION_DEBUG_FS 1577 bool "Debugfs entries for fault-injection capabilities" 1578 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1579 help 1580 Enable configuration of fault-injection capabilities via debugfs. 1581 1582config FAULT_INJECTION_STACKTRACE_FILTER 1583 bool "stacktrace filter for fault-injection capabilities" 1584 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1585 depends on !X86_64 1586 select STACKTRACE 1587 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !SCORE && !X86 1588 help 1589 Provide stacktrace filter for fault-injection capabilities 1590 1591config LATENCYTOP 1592 bool "Latency measuring infrastructure" 1593 depends on DEBUG_KERNEL 1594 depends on STACKTRACE_SUPPORT 1595 depends on PROC_FS 1596 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM_UNWIND && !ARC && !X86 1597 select KALLSYMS 1598 select KALLSYMS_ALL 1599 select STACKTRACE 1600 select SCHEDSTATS 1601 select SCHED_DEBUG 1602 help 1603 Enable this option if you want to use the LatencyTOP tool 1604 to find out which userspace is blocking on what kernel operations. 1605 1606source kernel/trace/Kconfig 1607 1608config PROVIDE_OHCI1394_DMA_INIT 1609 bool "Remote debugging over FireWire early on boot" 1610 depends on PCI && X86 1611 help 1612 If you want to debug problems which hang or crash the kernel early 1613 on boot and the crashing machine has a FireWire port, you can use 1614 this feature to remotely access the memory of the crashed machine 1615 over FireWire. This employs remote DMA as part of the OHCI1394 1616 specification which is now the standard for FireWire controllers. 1617 1618 With remote DMA, you can monitor the printk buffer remotely using 1619 firescope and access all memory below 4GB using fireproxy from gdb. 1620 Even controlling a kernel debugger is possible using remote DMA. 1621 1622 Usage: 1623 1624 If ohci1394_dma=early is used as boot parameter, it will initialize 1625 all OHCI1394 controllers which are found in the PCI config space. 1626 1627 As all changes to the FireWire bus such as enabling and disabling 1628 devices cause a bus reset and thereby disable remote DMA for all 1629 devices, be sure to have the cable plugged and FireWire enabled on 1630 the debugging host before booting the debug target for debugging. 1631 1632 This code (~1k) is freed after boot. By then, the firewire stack 1633 in charge of the OHCI-1394 controllers should be used instead. 1634 1635 See Documentation/debugging-via-ohci1394.txt for more information. 1636 1637config DMA_API_DEBUG 1638 bool "Enable debugging of DMA-API usage" 1639 depends on HAVE_DMA_API_DEBUG 1640 help 1641 Enable this option to debug the use of the DMA API by device drivers. 1642 With this option you will be able to detect common bugs in device 1643 drivers like double-freeing of DMA mappings or freeing mappings that 1644 were never allocated. 1645 1646 This also attempts to catch cases where a page owned by DMA is 1647 accessed by the cpu in a way that could cause data corruption. For 1648 example, this enables cow_user_page() to check that the source page is 1649 not undergoing DMA. 1650 1651 This option causes a performance degradation. Use only if you want to 1652 debug device drivers and dma interactions. 1653 1654 If unsure, say N. 1655 1656menu "Runtime Testing" 1657 1658config LKDTM 1659 tristate "Linux Kernel Dump Test Tool Module" 1660 depends on DEBUG_FS 1661 depends on BLOCK 1662 default n 1663 help 1664 This module enables testing of the different dumping mechanisms by 1665 inducing system failures at predefined crash points. 1666 If you don't need it: say N 1667 Choose M here to compile this code as a module. The module will be 1668 called lkdtm. 1669 1670 Documentation on how to use the module can be found in 1671 Documentation/fault-injection/provoke-crashes.txt 1672 1673config TEST_LIST_SORT 1674 tristate "Linked list sorting test" 1675 depends on DEBUG_KERNEL || m 1676 help 1677 Enable this to turn on 'list_sort()' function test. This test is 1678 executed only once during system boot (so affects only boot time), 1679 or at module load time. 1680 1681 If unsure, say N. 1682 1683config TEST_SORT 1684 tristate "Array-based sort test" 1685 depends on DEBUG_KERNEL || m 1686 help 1687 This option enables the self-test function of 'sort()' at boot, 1688 or at module load time. 1689 1690 If unsure, say N. 1691 1692config KPROBES_SANITY_TEST 1693 bool "Kprobes sanity tests" 1694 depends on DEBUG_KERNEL 1695 depends on KPROBES 1696 default n 1697 help 1698 This option provides for testing basic kprobes functionality on 1699 boot. A sample kprobe, jprobe and kretprobe are inserted and 1700 verified for functionality. 1701 1702 Say N if you are unsure. 1703 1704config BACKTRACE_SELF_TEST 1705 tristate "Self test for the backtrace code" 1706 depends on DEBUG_KERNEL 1707 default n 1708 help 1709 This option provides a kernel module that can be used to test 1710 the kernel stack backtrace code. This option is not useful 1711 for distributions or general kernels, but only for kernel 1712 developers working on architecture code. 1713 1714 Note that if you want to also test saved backtraces, you will 1715 have to enable STACKTRACE as well. 1716 1717 Say N if you are unsure. 1718 1719config RBTREE_TEST 1720 tristate "Red-Black tree test" 1721 depends on DEBUG_KERNEL 1722 help 1723 A benchmark measuring the performance of the rbtree library. 1724 Also includes rbtree invariant checks. 1725 1726config INTERVAL_TREE_TEST 1727 tristate "Interval tree test" 1728 depends on DEBUG_KERNEL 1729 select INTERVAL_TREE 1730 help 1731 A benchmark measuring the performance of the interval tree library 1732 1733config PERCPU_TEST 1734 tristate "Per cpu operations test" 1735 depends on m && DEBUG_KERNEL 1736 help 1737 Enable this option to build test module which validates per-cpu 1738 operations. 1739 1740 If unsure, say N. 1741 1742config ATOMIC64_SELFTEST 1743 tristate "Perform an atomic64_t self-test" 1744 help 1745 Enable this option to test the atomic64_t functions at boot or 1746 at module load time. 1747 1748 If unsure, say N. 1749 1750config ASYNC_RAID6_TEST 1751 tristate "Self test for hardware accelerated raid6 recovery" 1752 depends on ASYNC_RAID6_RECOV 1753 select ASYNC_MEMCPY 1754 ---help--- 1755 This is a one-shot self test that permutes through the 1756 recovery of all the possible two disk failure scenarios for a 1757 N-disk array. Recovery is performed with the asynchronous 1758 raid6 recovery routines, and will optionally use an offload 1759 engine if one is available. 1760 1761 If unsure, say N. 1762 1763config TEST_HEXDUMP 1764 tristate "Test functions located in the hexdump module at runtime" 1765 1766config TEST_STRING_HELPERS 1767 tristate "Test functions located in the string_helpers module at runtime" 1768 1769config TEST_KSTRTOX 1770 tristate "Test kstrto*() family of functions at runtime" 1771 1772config TEST_PRINTF 1773 tristate "Test printf() family of functions at runtime" 1774 1775config TEST_BITMAP 1776 tristate "Test bitmap_*() family of functions at runtime" 1777 default n 1778 help 1779 Enable this option to test the bitmap functions at boot. 1780 1781 If unsure, say N. 1782 1783config TEST_UUID 1784 tristate "Test functions located in the uuid module at runtime" 1785 1786config TEST_RHASHTABLE 1787 tristate "Perform selftest on resizable hash table" 1788 default n 1789 help 1790 Enable this option to test the rhashtable functions at boot. 1791 1792 If unsure, say N. 1793 1794config TEST_HASH 1795 tristate "Perform selftest on hash functions" 1796 default n 1797 help 1798 Enable this option to test the kernel's integer (<linux/hash.h>), 1799 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>) 1800 hash functions on boot (or module load). 1801 1802 This is intended to help people writing architecture-specific 1803 optimized versions. If unsure, say N. 1804 1805config TEST_PARMAN 1806 tristate "Perform selftest on priority array manager" 1807 default n 1808 depends on PARMAN 1809 help 1810 Enable this option to test priority array manager on boot 1811 (or module load). 1812 1813 If unsure, say N. 1814 1815config TEST_LKM 1816 tristate "Test module loading with 'hello world' module" 1817 default n 1818 depends on m 1819 help 1820 This builds the "test_module" module that emits "Hello, world" 1821 on printk when loaded. It is designed to be used for basic 1822 evaluation of the module loading subsystem (for example when 1823 validating module verification). It lacks any extra dependencies, 1824 and will not normally be loaded by the system unless explicitly 1825 requested by name. 1826 1827 If unsure, say N. 1828 1829config TEST_USER_COPY 1830 tristate "Test user/kernel boundary protections" 1831 default n 1832 depends on m 1833 help 1834 This builds the "test_user_copy" module that runs sanity checks 1835 on the copy_to/from_user infrastructure, making sure basic 1836 user/kernel boundary testing is working. If it fails to load, 1837 a regression has been detected in the user/kernel memory boundary 1838 protections. 1839 1840 If unsure, say N. 1841 1842config TEST_BPF 1843 tristate "Test BPF filter functionality" 1844 default n 1845 depends on m && NET 1846 help 1847 This builds the "test_bpf" module that runs various test vectors 1848 against the BPF interpreter or BPF JIT compiler depending on the 1849 current setting. This is in particular useful for BPF JIT compiler 1850 development, but also to run regression tests against changes in 1851 the interpreter code. It also enables test stubs for eBPF maps and 1852 verifier used by user space verifier testsuite. 1853 1854 If unsure, say N. 1855 1856config TEST_FIRMWARE 1857 tristate "Test firmware loading via userspace interface" 1858 default n 1859 depends on FW_LOADER 1860 help 1861 This builds the "test_firmware" module that creates a userspace 1862 interface for testing firmware loading. This can be used to 1863 control the triggering of firmware loading without needing an 1864 actual firmware-using device. The contents can be rechecked by 1865 userspace. 1866 1867 If unsure, say N. 1868 1869config TEST_SYSCTL 1870 tristate "sysctl test driver" 1871 default n 1872 depends on PROC_SYSCTL 1873 help 1874 This builds the "test_sysctl" module. This driver enables to test the 1875 proc sysctl interfaces available to drivers safely without affecting 1876 production knobs which might alter system functionality. 1877 1878 If unsure, say N. 1879 1880config TEST_UDELAY 1881 tristate "udelay test driver" 1882 default n 1883 help 1884 This builds the "udelay_test" module that helps to make sure 1885 that udelay() is working properly. 1886 1887 If unsure, say N. 1888 1889config TEST_STATIC_KEYS 1890 tristate "Test static keys" 1891 default n 1892 depends on m 1893 help 1894 Test the static key interfaces. 1895 1896 If unsure, say N. 1897 1898config TEST_KMOD 1899 tristate "kmod stress tester" 1900 default n 1901 depends on m 1902 depends on BLOCK && (64BIT || LBDAF) # for XFS, BTRFS 1903 depends on NETDEVICES && NET_CORE && INET # for TUN 1904 select TEST_LKM 1905 select XFS_FS 1906 select TUN 1907 select BTRFS_FS 1908 help 1909 Test the kernel's module loading mechanism: kmod. kmod implements 1910 support to load modules using the Linux kernel's usermode helper. 1911 This test provides a series of tests against kmod. 1912 1913 Although technically you can either build test_kmod as a module or 1914 into the kernel we disallow building it into the kernel since 1915 it stress tests request_module() and this will very likely cause 1916 some issues by taking over precious threads available from other 1917 module load requests, ultimately this could be fatal. 1918 1919 To run tests run: 1920 1921 tools/testing/selftests/kmod/kmod.sh --help 1922 1923 If unsure, say N. 1924 1925config TEST_DEBUG_VIRTUAL 1926 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 1927 depends on DEBUG_VIRTUAL 1928 help 1929 Test the kernel's ability to detect incorrect calls to 1930 virt_to_phys() done against the non-linear part of the 1931 kernel's virtual address map. 1932 1933 If unsure, say N. 1934 1935endmenu # runtime tests 1936 1937config MEMTEST 1938 bool "Memtest" 1939 depends on HAVE_MEMBLOCK 1940 ---help--- 1941 This option adds a kernel parameter 'memtest', which allows memtest 1942 to be set. 1943 memtest=0, mean disabled; -- default 1944 memtest=1, mean do 1 test pattern; 1945 ... 1946 memtest=17, mean do 17 test patterns. 1947 If you are unsure how to answer this question, answer N. 1948 1949config BUG_ON_DATA_CORRUPTION 1950 bool "Trigger a BUG when data corruption is detected" 1951 select DEBUG_LIST 1952 help 1953 Select this option if the kernel should BUG when it encounters 1954 data corruption in kernel memory structures when they get checked 1955 for validity. 1956 1957 If unsure, say N. 1958 1959source "samples/Kconfig" 1960 1961source "lib/Kconfig.kgdb" 1962 1963source "lib/Kconfig.ubsan" 1964 1965config ARCH_HAS_DEVMEM_IS_ALLOWED 1966 bool 1967 1968config STRICT_DEVMEM 1969 bool "Filter access to /dev/mem" 1970 depends on MMU && DEVMEM 1971 depends on ARCH_HAS_DEVMEM_IS_ALLOWED 1972 default y if TILE || PPC 1973 ---help--- 1974 If this option is disabled, you allow userspace (root) access to all 1975 of memory, including kernel and userspace memory. Accidental 1976 access to this is obviously disastrous, but specific access can 1977 be used by people debugging the kernel. Note that with PAT support 1978 enabled, even in this case there are restrictions on /dev/mem 1979 use due to the cache aliasing requirements. 1980 1981 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem 1982 file only allows userspace access to PCI space and the BIOS code and 1983 data regions. This is sufficient for dosemu and X and all common 1984 users of /dev/mem. 1985 1986 If in doubt, say Y. 1987 1988config IO_STRICT_DEVMEM 1989 bool "Filter I/O access to /dev/mem" 1990 depends on STRICT_DEVMEM 1991 ---help--- 1992 If this option is disabled, you allow userspace (root) access to all 1993 io-memory regardless of whether a driver is actively using that 1994 range. Accidental access to this is obviously disastrous, but 1995 specific access can be used by people debugging kernel drivers. 1996 1997 If this option is switched on, the /dev/mem file only allows 1998 userspace access to *idle* io-memory ranges (see /proc/iomem) This 1999 may break traditional users of /dev/mem (dosemu, legacy X, etc...) 2000 if the driver using a given range cannot be disabled. 2001 2002 If in doubt, say Y. 2003