1# SPDX-License-Identifier: GPL-2.0-only 2menu "Kernel hacking" 3 4menu "printk and dmesg options" 5 6config PRINTK_TIME 7 bool "Show timing information on printks" 8 depends on PRINTK 9 help 10 Selecting this option causes time stamps of the printk() 11 messages to be added to the output of the syslog() system 12 call and at the console. 13 14 The timestamp is always recorded internally, and exported 15 to /dev/kmsg. This flag just specifies if the timestamp should 16 be included, not that the timestamp is recorded. 17 18 The behavior is also controlled by the kernel command line 19 parameter printk.time=1. See Documentation/admin-guide/kernel-parameters.rst 20 21config PRINTK_CALLER 22 bool "Show caller information on printks" 23 depends on PRINTK 24 help 25 Selecting this option causes printk() to add a caller "thread id" (if 26 in task context) or a caller "processor id" (if not in task context) 27 to every message. 28 29 This option is intended for environments where multiple threads 30 concurrently call printk() for many times, for it is difficult to 31 interpret without knowing where these lines (or sometimes individual 32 line which was divided into multiple lines due to race) came from. 33 34 Since toggling after boot makes the code racy, currently there is 35 no option to enable/disable at the kernel command line parameter or 36 sysfs interface. 37 38config CONSOLE_LOGLEVEL_DEFAULT 39 int "Default console loglevel (1-15)" 40 range 1 15 41 default "7" 42 help 43 Default loglevel to determine what will be printed on the console. 44 45 Setting a default here is equivalent to passing in loglevel=<x> in 46 the kernel bootargs. loglevel=<x> continues to override whatever 47 value is specified here as well. 48 49 Note: This does not affect the log level of un-prefixed printk() 50 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT 51 option. 52 53config CONSOLE_LOGLEVEL_QUIET 54 int "quiet console loglevel (1-15)" 55 range 1 15 56 default "4" 57 help 58 loglevel to use when "quiet" is passed on the kernel commandline. 59 60 When "quiet" is passed on the kernel commandline this loglevel 61 will be used as the loglevel. IOW passing "quiet" will be the 62 equivalent of passing "loglevel=<CONSOLE_LOGLEVEL_QUIET>" 63 64config MESSAGE_LOGLEVEL_DEFAULT 65 int "Default message log level (1-7)" 66 range 1 7 67 default "4" 68 help 69 Default log level for printk statements with no specified priority. 70 71 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks 72 that are auditing their logs closely may want to set it to a lower 73 priority. 74 75 Note: This does not affect what message level gets printed on the console 76 by default. To change that, use loglevel=<x> in the kernel bootargs, 77 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value. 78 79config BOOT_PRINTK_DELAY 80 bool "Delay each boot printk message by N milliseconds" 81 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY 82 help 83 This build option allows you to read kernel boot messages 84 by inserting a short delay after each one. The delay is 85 specified in milliseconds on the kernel command line, 86 using "boot_delay=N". 87 88 It is likely that you would also need to use "lpj=M" to preset 89 the "loops per jiffie" value. 90 See a previous boot log for the "lpj" value to use for your 91 system, and then set "lpj=M" before setting "boot_delay=N". 92 NOTE: Using this option may adversely affect SMP systems. 93 I.e., processors other than the first one may not boot up. 94 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect 95 what it believes to be lockup conditions. 96 97config DYNAMIC_DEBUG 98 bool "Enable dynamic printk() support" 99 default n 100 depends on PRINTK 101 depends on (DEBUG_FS || PROC_FS) 102 select DYNAMIC_DEBUG_CORE 103 help 104 105 Compiles debug level messages into the kernel, which would not 106 otherwise be available at runtime. These messages can then be 107 enabled/disabled based on various levels of scope - per source file, 108 function, module, format string, and line number. This mechanism 109 implicitly compiles in all pr_debug() and dev_dbg() calls, which 110 enlarges the kernel text size by about 2%. 111 112 If a source file is compiled with DEBUG flag set, any 113 pr_debug() calls in it are enabled by default, but can be 114 disabled at runtime as below. Note that DEBUG flag is 115 turned on by many CONFIG_*DEBUG* options. 116 117 Usage: 118 119 Dynamic debugging is controlled via the 'dynamic_debug/control' file, 120 which is contained in the 'debugfs' filesystem or procfs. 121 Thus, the debugfs or procfs filesystem must first be mounted before 122 making use of this feature. 123 We refer the control file as: <debugfs>/dynamic_debug/control. This 124 file contains a list of the debug statements that can be enabled. The 125 format for each line of the file is: 126 127 filename:lineno [module]function flags format 128 129 filename : source file of the debug statement 130 lineno : line number of the debug statement 131 module : module that contains the debug statement 132 function : function that contains the debug statement 133 flags : '=p' means the line is turned 'on' for printing 134 format : the format used for the debug statement 135 136 From a live system: 137 138 nullarbor:~ # cat <debugfs>/dynamic_debug/control 139 # filename:lineno [module]function flags format 140 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012" 141 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012" 142 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012" 143 144 Example usage: 145 146 // enable the message at line 1603 of file svcsock.c 147 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > 148 <debugfs>/dynamic_debug/control 149 150 // enable all the messages in file svcsock.c 151 nullarbor:~ # echo -n 'file svcsock.c +p' > 152 <debugfs>/dynamic_debug/control 153 154 // enable all the messages in the NFS server module 155 nullarbor:~ # echo -n 'module nfsd +p' > 156 <debugfs>/dynamic_debug/control 157 158 // enable all 12 messages in the function svc_process() 159 nullarbor:~ # echo -n 'func svc_process +p' > 160 <debugfs>/dynamic_debug/control 161 162 // disable all 12 messages in the function svc_process() 163 nullarbor:~ # echo -n 'func svc_process -p' > 164 <debugfs>/dynamic_debug/control 165 166 See Documentation/admin-guide/dynamic-debug-howto.rst for additional 167 information. 168 169config DYNAMIC_DEBUG_CORE 170 bool "Enable core function of dynamic debug support" 171 depends on PRINTK 172 depends on (DEBUG_FS || PROC_FS) 173 help 174 Enable core functional support of dynamic debug. It is useful 175 when you want to tie dynamic debug to your kernel modules with 176 DYNAMIC_DEBUG_MODULE defined for each of them, especially for 177 the case of embedded system where the kernel image size is 178 sensitive for people. 179 180config SYMBOLIC_ERRNAME 181 bool "Support symbolic error names in printf" 182 default y if PRINTK 183 help 184 If you say Y here, the kernel's printf implementation will 185 be able to print symbolic error names such as ENOSPC instead 186 of the number 28. It makes the kernel image slightly larger 187 (about 3KB), but can make the kernel logs easier to read. 188 189config DEBUG_BUGVERBOSE 190 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 191 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE) 192 default y 193 help 194 Say Y here to make BUG() panics output the file name and line number 195 of the BUG call as well as the EIP and oops trace. This aids 196 debugging but costs about 70-100K of memory. 197 198endmenu # "printk and dmesg options" 199 200menu "Compile-time checks and compiler options" 201 202config DEBUG_INFO 203 bool "Compile the kernel with debug info" 204 depends on DEBUG_KERNEL && !COMPILE_TEST 205 help 206 If you say Y here the resulting kernel image will include 207 debugging info resulting in a larger kernel image. 208 This adds debug symbols to the kernel and modules (gcc -g), and 209 is needed if you intend to use kernel crashdump or binary object 210 tools like crash, kgdb, LKCD, gdb, etc on the kernel. 211 Say Y here only if you plan to debug the kernel. 212 213 If unsure, say N. 214 215if DEBUG_INFO 216 217config DEBUG_INFO_REDUCED 218 bool "Reduce debugging information" 219 help 220 If you say Y here gcc is instructed to generate less debugging 221 information for structure types. This means that tools that 222 need full debugging information (like kgdb or systemtap) won't 223 be happy. But if you merely need debugging information to 224 resolve line numbers there is no loss. Advantage is that 225 build directory object sizes shrink dramatically over a full 226 DEBUG_INFO build and compile times are reduced too. 227 Only works with newer gcc versions. 228 229config DEBUG_INFO_COMPRESSED 230 bool "Compressed debugging information" 231 depends on $(cc-option,-gz=zlib) 232 depends on $(ld-option,--compress-debug-sections=zlib) 233 help 234 Compress the debug information using zlib. Requires GCC 5.0+ or Clang 235 5.0+, binutils 2.26+, and zlib. 236 237 Users of dpkg-deb via scripts/package/builddeb may find an increase in 238 size of their debug .deb packages with this config set, due to the 239 debug info being compressed with zlib, then the object files being 240 recompressed with a different compression scheme. But this is still 241 preferable to setting $KDEB_COMPRESS to "none" which would be even 242 larger. 243 244config DEBUG_INFO_SPLIT 245 bool "Produce split debuginfo in .dwo files" 246 depends on $(cc-option,-gsplit-dwarf) 247 help 248 Generate debug info into separate .dwo files. This significantly 249 reduces the build directory size for builds with DEBUG_INFO, 250 because it stores the information only once on disk in .dwo 251 files instead of multiple times in object files and executables. 252 In addition the debug information is also compressed. 253 254 Requires recent gcc (4.7+) and recent gdb/binutils. 255 Any tool that packages or reads debug information would need 256 to know about the .dwo files and include them. 257 Incompatible with older versions of ccache. 258 259config DEBUG_INFO_DWARF4 260 bool "Generate dwarf4 debuginfo" 261 depends on $(cc-option,-gdwarf-4) 262 help 263 Generate dwarf4 debug info. This requires recent versions 264 of gcc and gdb. It makes the debug information larger. 265 But it significantly improves the success of resolving 266 variables in gdb on optimized code. 267 268config DEBUG_INFO_BTF 269 bool "Generate BTF typeinfo" 270 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED 271 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST 272 help 273 Generate deduplicated BTF type information from DWARF debug info. 274 Turning this on expects presence of pahole tool, which will convert 275 DWARF type info into equivalent deduplicated BTF type info. 276 277config GDB_SCRIPTS 278 bool "Provide GDB scripts for kernel debugging" 279 help 280 This creates the required links to GDB helper scripts in the 281 build directory. If you load vmlinux into gdb, the helper 282 scripts will be automatically imported by gdb as well, and 283 additional functions are available to analyze a Linux kernel 284 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst 285 for further details. 286 287endif # DEBUG_INFO 288 289config ENABLE_MUST_CHECK 290 bool "Enable __must_check logic" 291 default y 292 help 293 Enable the __must_check logic in the kernel build. Disable this to 294 suppress the "warning: ignoring return value of 'foo', declared with 295 attribute warn_unused_result" messages. 296 297config FRAME_WARN 298 int "Warn for stack frames larger than" 299 range 0 8192 300 default 2048 if GCC_PLUGIN_LATENT_ENTROPY 301 default 1280 if (!64BIT && PARISC) 302 default 1024 if (!64BIT && !PARISC) 303 default 2048 if 64BIT 304 help 305 Tell gcc to warn at build time for stack frames larger than this. 306 Setting this too low will cause a lot of warnings. 307 Setting it to 0 disables the warning. 308 309config STRIP_ASM_SYMS 310 bool "Strip assembler-generated symbols during link" 311 default n 312 help 313 Strip internal assembler-generated symbols during a link (symbols 314 that look like '.Lxxx') so they don't pollute the output of 315 get_wchan() and suchlike. 316 317config READABLE_ASM 318 bool "Generate readable assembler code" 319 depends on DEBUG_KERNEL 320 help 321 Disable some compiler optimizations that tend to generate human unreadable 322 assembler output. This may make the kernel slightly slower, but it helps 323 to keep kernel developers who have to stare a lot at assembler listings 324 sane. 325 326config HEADERS_INSTALL 327 bool "Install uapi headers to usr/include" 328 depends on !UML 329 help 330 This option will install uapi headers (headers exported to user-space) 331 into the usr/include directory for use during the kernel build. 332 This is unneeded for building the kernel itself, but needed for some 333 user-space program samples. It is also needed by some features such 334 as uapi header sanity checks. 335 336config DEBUG_SECTION_MISMATCH 337 bool "Enable full Section mismatch analysis" 338 help 339 The section mismatch analysis checks if there are illegal 340 references from one section to another section. 341 During linktime or runtime, some sections are dropped; 342 any use of code/data previously in these sections would 343 most likely result in an oops. 344 In the code, functions and variables are annotated with 345 __init,, etc. (see the full list in include/linux/init.h), 346 which results in the code/data being placed in specific sections. 347 The section mismatch analysis is always performed after a full 348 kernel build, and enabling this option causes the following 349 additional step to occur: 350 - Add the option -fno-inline-functions-called-once to gcc commands. 351 When inlining a function annotated with __init in a non-init 352 function, we would lose the section information and thus 353 the analysis would not catch the illegal reference. 354 This option tells gcc to inline less (but it does result in 355 a larger kernel). 356 357config SECTION_MISMATCH_WARN_ONLY 358 bool "Make section mismatch errors non-fatal" 359 default y 360 help 361 If you say N here, the build process will fail if there are any 362 section mismatch, instead of just throwing warnings. 363 364 If unsure, say Y. 365 366config DEBUG_FORCE_FUNCTION_ALIGN_32B 367 bool "Force all function address 32B aligned" if EXPERT 368 help 369 There are cases that a commit from one domain changes the function 370 address alignment of other domains, and cause magic performance 371 bump (regression or improvement). Enable this option will help to 372 verify if the bump is caused by function alignment changes, while 373 it will slightly increase the kernel size and affect icache usage. 374 375 It is mainly for debug and performance tuning use. 376 377# 378# Select this config option from the architecture Kconfig, if it 379# is preferred to always offer frame pointers as a config 380# option on the architecture (regardless of KERNEL_DEBUG): 381# 382config ARCH_WANT_FRAME_POINTERS 383 bool 384 385config FRAME_POINTER 386 bool "Compile the kernel with frame pointers" 387 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS 388 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 389 help 390 If you say Y here the resulting kernel image will be slightly 391 larger and slower, but it gives very useful debugging information 392 in case of kernel bugs. (precise oopses/stacktraces/warnings) 393 394config STACK_VALIDATION 395 bool "Compile-time stack metadata validation" 396 depends on HAVE_STACK_VALIDATION 397 default n 398 help 399 Add compile-time checks to validate stack metadata, including frame 400 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure 401 that runtime stack traces are more reliable. 402 403 This is also a prerequisite for generation of ORC unwind data, which 404 is needed for CONFIG_UNWINDER_ORC. 405 406 For more information, see 407 tools/objtool/Documentation/stack-validation.txt. 408 409config VMLINUX_VALIDATION 410 bool 411 depends on STACK_VALIDATION && DEBUG_ENTRY && !PARAVIRT 412 default y 413 414config DEBUG_FORCE_WEAK_PER_CPU 415 bool "Force weak per-cpu definitions" 416 depends on DEBUG_KERNEL 417 help 418 s390 and alpha require percpu variables in modules to be 419 defined weak to work around addressing range issue which 420 puts the following two restrictions on percpu variable 421 definitions. 422 423 1. percpu symbols must be unique whether static or not 424 2. percpu variables can't be defined inside a function 425 426 To ensure that generic code follows the above rules, this 427 option forces all percpu variables to be defined as weak. 428 429endmenu # "Compiler options" 430 431menu "Generic Kernel Debugging Instruments" 432 433config MAGIC_SYSRQ 434 bool "Magic SysRq key" 435 depends on !UML 436 help 437 If you say Y here, you will have some control over the system even 438 if the system crashes for example during kernel debugging (e.g., you 439 will be able to flush the buffer cache to disk, reboot the system 440 immediately or dump some status information). This is accomplished 441 by pressing various keys while holding SysRq (Alt+PrintScreen). It 442 also works on a serial console (on PC hardware at least), if you 443 send a BREAK and then within 5 seconds a command keypress. The 444 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 445 Don't say Y unless you really know what this hack does. 446 447config MAGIC_SYSRQ_DEFAULT_ENABLE 448 hex "Enable magic SysRq key functions by default" 449 depends on MAGIC_SYSRQ 450 default 0x1 451 help 452 Specifies which SysRq key functions are enabled by default. 453 This may be set to 1 or 0 to enable or disable them all, or 454 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 455 456config MAGIC_SYSRQ_SERIAL 457 bool "Enable magic SysRq key over serial" 458 depends on MAGIC_SYSRQ 459 default y 460 help 461 Many embedded boards have a disconnected TTL level serial which can 462 generate some garbage that can lead to spurious false sysrq detects. 463 This option allows you to decide whether you want to enable the 464 magic SysRq key. 465 466config MAGIC_SYSRQ_SERIAL_SEQUENCE 467 string "Char sequence that enables magic SysRq over serial" 468 depends on MAGIC_SYSRQ_SERIAL 469 default "" 470 help 471 Specifies a sequence of characters that can follow BREAK to enable 472 SysRq on a serial console. 473 474 If unsure, leave an empty string and the option will not be enabled. 475 476config DEBUG_FS 477 bool "Debug Filesystem" 478 help 479 debugfs is a virtual file system that kernel developers use to put 480 debugging files into. Enable this option to be able to read and 481 write to these files. 482 483 For detailed documentation on the debugfs API, see 484 Documentation/filesystems/. 485 486 If unsure, say N. 487 488choice 489 prompt "Debugfs default access" 490 depends on DEBUG_FS 491 default DEBUG_FS_ALLOW_ALL 492 help 493 This selects the default access restrictions for debugfs. 494 It can be overridden with kernel command line option 495 debugfs=[on,no-mount,off]. The restrictions apply for API access 496 and filesystem registration. 497 498config DEBUG_FS_ALLOW_ALL 499 bool "Access normal" 500 help 501 No restrictions apply. Both API and filesystem registration 502 is on. This is the normal default operation. 503 504config DEBUG_FS_DISALLOW_MOUNT 505 bool "Do not register debugfs as filesystem" 506 help 507 The API is open but filesystem is not loaded. Clients can still do 508 their work and read with debug tools that do not need 509 debugfs filesystem. 510 511config DEBUG_FS_ALLOW_NONE 512 bool "No access" 513 help 514 Access is off. Clients get -PERM when trying to create nodes in 515 debugfs tree and debugfs is not registered as a filesystem. 516 Client can then back-off or continue without debugfs access. 517 518endchoice 519 520source "lib/Kconfig.kgdb" 521 522source "lib/Kconfig.ubsan" 523 524endmenu 525 526config DEBUG_KERNEL 527 bool "Kernel debugging" 528 help 529 Say Y here if you are developing drivers or trying to debug and 530 identify kernel problems. 531 532config DEBUG_MISC 533 bool "Miscellaneous debug code" 534 default DEBUG_KERNEL 535 depends on DEBUG_KERNEL 536 help 537 Say Y here if you need to enable miscellaneous debug code that should 538 be under a more specific debug option but isn't. 539 540 541menu "Memory Debugging" 542 543source "mm/Kconfig.debug" 544 545config DEBUG_OBJECTS 546 bool "Debug object operations" 547 depends on DEBUG_KERNEL 548 help 549 If you say Y here, additional code will be inserted into the 550 kernel to track the life time of various objects and validate 551 the operations on those objects. 552 553config DEBUG_OBJECTS_SELFTEST 554 bool "Debug objects selftest" 555 depends on DEBUG_OBJECTS 556 help 557 This enables the selftest of the object debug code. 558 559config DEBUG_OBJECTS_FREE 560 bool "Debug objects in freed memory" 561 depends on DEBUG_OBJECTS 562 help 563 This enables checks whether a k/v free operation frees an area 564 which contains an object which has not been deactivated 565 properly. This can make kmalloc/kfree-intensive workloads 566 much slower. 567 568config DEBUG_OBJECTS_TIMERS 569 bool "Debug timer objects" 570 depends on DEBUG_OBJECTS 571 help 572 If you say Y here, additional code will be inserted into the 573 timer routines to track the life time of timer objects and 574 validate the timer operations. 575 576config DEBUG_OBJECTS_WORK 577 bool "Debug work objects" 578 depends on DEBUG_OBJECTS 579 help 580 If you say Y here, additional code will be inserted into the 581 work queue routines to track the life time of work objects and 582 validate the work operations. 583 584config DEBUG_OBJECTS_RCU_HEAD 585 bool "Debug RCU callbacks objects" 586 depends on DEBUG_OBJECTS 587 help 588 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 589 590config DEBUG_OBJECTS_PERCPU_COUNTER 591 bool "Debug percpu counter objects" 592 depends on DEBUG_OBJECTS 593 help 594 If you say Y here, additional code will be inserted into the 595 percpu counter routines to track the life time of percpu counter 596 objects and validate the percpu counter operations. 597 598config DEBUG_OBJECTS_ENABLE_DEFAULT 599 int "debug_objects bootup default value (0-1)" 600 range 0 1 601 default "1" 602 depends on DEBUG_OBJECTS 603 help 604 Debug objects boot parameter default value 605 606config DEBUG_SLAB 607 bool "Debug slab memory allocations" 608 depends on DEBUG_KERNEL && SLAB 609 help 610 Say Y here to have the kernel do limited verification on memory 611 allocation as well as poisoning memory on free to catch use of freed 612 memory. This can make kmalloc/kfree-intensive workloads much slower. 613 614config SLUB_DEBUG_ON 615 bool "SLUB debugging on by default" 616 depends on SLUB && SLUB_DEBUG 617 default n 618 help 619 Boot with debugging on by default. SLUB boots by default with 620 the runtime debug capabilities switched off. Enabling this is 621 equivalent to specifying the "slub_debug" parameter on boot. 622 There is no support for more fine grained debug control like 623 possible with slub_debug=xxx. SLUB debugging may be switched 624 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 625 "slub_debug=-". 626 627config SLUB_STATS 628 default n 629 bool "Enable SLUB performance statistics" 630 depends on SLUB && SYSFS 631 help 632 SLUB statistics are useful to debug SLUBs allocation behavior in 633 order find ways to optimize the allocator. This should never be 634 enabled for production use since keeping statistics slows down 635 the allocator by a few percentage points. The slabinfo command 636 supports the determination of the most active slabs to figure 637 out which slabs are relevant to a particular load. 638 Try running: slabinfo -DA 639 640config HAVE_DEBUG_KMEMLEAK 641 bool 642 643config DEBUG_KMEMLEAK 644 bool "Kernel memory leak detector" 645 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 646 select DEBUG_FS 647 select STACKTRACE if STACKTRACE_SUPPORT 648 select KALLSYMS 649 select CRC32 650 help 651 Say Y here if you want to enable the memory leak 652 detector. The memory allocation/freeing is traced in a way 653 similar to the Boehm's conservative garbage collector, the 654 difference being that the orphan objects are not freed but 655 only shown in /sys/kernel/debug/kmemleak. Enabling this 656 feature will introduce an overhead to memory 657 allocations. See Documentation/dev-tools/kmemleak.rst for more 658 details. 659 660 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 661 of finding leaks due to the slab objects poisoning. 662 663 In order to access the kmemleak file, debugfs needs to be 664 mounted (usually at /sys/kernel/debug). 665 666config DEBUG_KMEMLEAK_MEM_POOL_SIZE 667 int "Kmemleak memory pool size" 668 depends on DEBUG_KMEMLEAK 669 range 200 1000000 670 default 16000 671 help 672 Kmemleak must track all the memory allocations to avoid 673 reporting false positives. Since memory may be allocated or 674 freed before kmemleak is fully initialised, use a static pool 675 of metadata objects to track such callbacks. After kmemleak is 676 fully initialised, this memory pool acts as an emergency one 677 if slab allocations fail. 678 679config DEBUG_KMEMLEAK_TEST 680 tristate "Simple test for the kernel memory leak detector" 681 depends on DEBUG_KMEMLEAK && m 682 help 683 This option enables a module that explicitly leaks memory. 684 685 If unsure, say N. 686 687config DEBUG_KMEMLEAK_DEFAULT_OFF 688 bool "Default kmemleak to off" 689 depends on DEBUG_KMEMLEAK 690 help 691 Say Y here to disable kmemleak by default. It can then be enabled 692 on the command line via kmemleak=on. 693 694config DEBUG_KMEMLEAK_AUTO_SCAN 695 bool "Enable kmemleak auto scan thread on boot up" 696 default y 697 depends on DEBUG_KMEMLEAK 698 help 699 Depending on the cpu, kmemleak scan may be cpu intensive and can 700 stall user tasks at times. This option enables/disables automatic 701 kmemleak scan at boot up. 702 703 Say N here to disable kmemleak auto scan thread to stop automatic 704 scanning. Disabling this option disables automatic reporting of 705 memory leaks. 706 707 If unsure, say Y. 708 709config DEBUG_STACK_USAGE 710 bool "Stack utilization instrumentation" 711 depends on DEBUG_KERNEL && !IA64 712 help 713 Enables the display of the minimum amount of free stack which each 714 task has ever had available in the sysrq-T and sysrq-P debug output. 715 716 This option will slow down process creation somewhat. 717 718config SCHED_STACK_END_CHECK 719 bool "Detect stack corruption on calls to schedule()" 720 depends on DEBUG_KERNEL 721 default n 722 help 723 This option checks for a stack overrun on calls to schedule(). 724 If the stack end location is found to be over written always panic as 725 the content of the corrupted region can no longer be trusted. 726 This is to ensure no erroneous behaviour occurs which could result in 727 data corruption or a sporadic crash at a later stage once the region 728 is examined. The runtime overhead introduced is minimal. 729 730config ARCH_HAS_DEBUG_VM_PGTABLE 731 bool 732 help 733 An architecture should select this when it can successfully 734 build and run DEBUG_VM_PGTABLE. 735 736config DEBUG_VM 737 bool "Debug VM" 738 depends on DEBUG_KERNEL 739 help 740 Enable this to turn on extended checks in the virtual-memory system 741 that may impact performance. 742 743 If unsure, say N. 744 745config DEBUG_VM_VMACACHE 746 bool "Debug VMA caching" 747 depends on DEBUG_VM 748 help 749 Enable this to turn on VMA caching debug information. Doing so 750 can cause significant overhead, so only enable it in non-production 751 environments. 752 753 If unsure, say N. 754 755config DEBUG_VM_RB 756 bool "Debug VM red-black trees" 757 depends on DEBUG_VM 758 help 759 Enable VM red-black tree debugging information and extra validations. 760 761 If unsure, say N. 762 763config DEBUG_VM_PGFLAGS 764 bool "Debug page-flags operations" 765 depends on DEBUG_VM 766 help 767 Enables extra validation on page flags operations. 768 769 If unsure, say N. 770 771config DEBUG_VM_PGTABLE 772 bool "Debug arch page table for semantics compliance" 773 depends on MMU 774 depends on ARCH_HAS_DEBUG_VM_PGTABLE 775 default y if DEBUG_VM 776 help 777 This option provides a debug method which can be used to test 778 architecture page table helper functions on various platforms in 779 verifying if they comply with expected generic MM semantics. This 780 will help architecture code in making sure that any changes or 781 new additions of these helpers still conform to expected 782 semantics of the generic MM. Platforms will have to opt in for 783 this through ARCH_HAS_DEBUG_VM_PGTABLE. 784 785 If unsure, say N. 786 787config ARCH_HAS_DEBUG_VIRTUAL 788 bool 789 790config DEBUG_VIRTUAL 791 bool "Debug VM translations" 792 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 793 help 794 Enable some costly sanity checks in virtual to page code. This can 795 catch mistakes with virt_to_page() and friends. 796 797 If unsure, say N. 798 799config DEBUG_NOMMU_REGIONS 800 bool "Debug the global anon/private NOMMU mapping region tree" 801 depends on DEBUG_KERNEL && !MMU 802 help 803 This option causes the global tree of anonymous and private mapping 804 regions to be regularly checked for invalid topology. 805 806config DEBUG_MEMORY_INIT 807 bool "Debug memory initialisation" if EXPERT 808 default !EXPERT 809 help 810 Enable this for additional checks during memory initialisation. 811 The sanity checks verify aspects of the VM such as the memory model 812 and other information provided by the architecture. Verbose 813 information will be printed at KERN_DEBUG loglevel depending 814 on the mminit_loglevel= command-line option. 815 816 If unsure, say Y 817 818config MEMORY_NOTIFIER_ERROR_INJECT 819 tristate "Memory hotplug notifier error injection module" 820 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 821 help 822 This option provides the ability to inject artificial errors to 823 memory hotplug notifier chain callbacks. It is controlled through 824 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 825 826 If the notifier call chain should be failed with some events 827 notified, write the error code to "actions/<notifier event>/error". 828 829 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 830 831 # cd /sys/kernel/debug/notifier-error-inject/memory 832 # echo -12 > actions/MEM_GOING_OFFLINE/error 833 # echo offline > /sys/devices/system/memory/memoryXXX/state 834 bash: echo: write error: Cannot allocate memory 835 836 To compile this code as a module, choose M here: the module will 837 be called memory-notifier-error-inject. 838 839 If unsure, say N. 840 841config DEBUG_PER_CPU_MAPS 842 bool "Debug access to per_cpu maps" 843 depends on DEBUG_KERNEL 844 depends on SMP 845 help 846 Say Y to verify that the per_cpu map being accessed has 847 been set up. This adds a fair amount of code to kernel memory 848 and decreases performance. 849 850 Say N if unsure. 851 852config DEBUG_HIGHMEM 853 bool "Highmem debugging" 854 depends on DEBUG_KERNEL && HIGHMEM 855 help 856 This option enables additional error checking for high memory 857 systems. Disable for production systems. 858 859config HAVE_DEBUG_STACKOVERFLOW 860 bool 861 862config DEBUG_STACKOVERFLOW 863 bool "Check for stack overflows" 864 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 865 help 866 Say Y here if you want to check for overflows of kernel, IRQ 867 and exception stacks (if your architecture uses them). This 868 option will show detailed messages if free stack space drops 869 below a certain limit. 870 871 These kinds of bugs usually occur when call-chains in the 872 kernel get too deep, especially when interrupts are 873 involved. 874 875 Use this in cases where you see apparently random memory 876 corruption, especially if it appears in 'struct thread_info' 877 878 If in doubt, say "N". 879 880source "lib/Kconfig.kasan" 881 882endmenu # "Memory Debugging" 883 884config DEBUG_SHIRQ 885 bool "Debug shared IRQ handlers" 886 depends on DEBUG_KERNEL 887 help 888 Enable this to generate a spurious interrupt just before a shared 889 interrupt handler is deregistered (generating one when registering 890 is currently disabled). Drivers need to handle this correctly. Some 891 don't and need to be caught. 892 893menu "Debug Oops, Lockups and Hangs" 894 895config PANIC_ON_OOPS 896 bool "Panic on Oops" 897 help 898 Say Y here to enable the kernel to panic when it oopses. This 899 has the same effect as setting oops=panic on the kernel command 900 line. 901 902 This feature is useful to ensure that the kernel does not do 903 anything erroneous after an oops which could result in data 904 corruption or other issues. 905 906 Say N if unsure. 907 908config PANIC_ON_OOPS_VALUE 909 int 910 range 0 1 911 default 0 if !PANIC_ON_OOPS 912 default 1 if PANIC_ON_OOPS 913 914config PANIC_TIMEOUT 915 int "panic timeout" 916 default 0 917 help 918 Set the timeout value (in seconds) until a reboot occurs when 919 the kernel panics. If n = 0, then we wait forever. A timeout 920 value n > 0 will wait n seconds before rebooting, while a timeout 921 value n < 0 will reboot immediately. 922 923config LOCKUP_DETECTOR 924 bool 925 926config SOFTLOCKUP_DETECTOR 927 bool "Detect Soft Lockups" 928 depends on DEBUG_KERNEL && !S390 929 select LOCKUP_DETECTOR 930 help 931 Say Y here to enable the kernel to act as a watchdog to detect 932 soft lockups. 933 934 Softlockups are bugs that cause the kernel to loop in kernel 935 mode for more than 20 seconds, without giving other tasks a 936 chance to run. The current stack trace is displayed upon 937 detection and the system will stay locked up. 938 939config BOOTPARAM_SOFTLOCKUP_PANIC 940 bool "Panic (Reboot) On Soft Lockups" 941 depends on SOFTLOCKUP_DETECTOR 942 help 943 Say Y here to enable the kernel to panic on "soft lockups", 944 which are bugs that cause the kernel to loop in kernel 945 mode for more than 20 seconds (configurable using the watchdog_thresh 946 sysctl), without giving other tasks a chance to run. 947 948 The panic can be used in combination with panic_timeout, 949 to cause the system to reboot automatically after a 950 lockup has been detected. This feature is useful for 951 high-availability systems that have uptime guarantees and 952 where a lockup must be resolved ASAP. 953 954 Say N if unsure. 955 956config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 957 int 958 depends on SOFTLOCKUP_DETECTOR 959 range 0 1 960 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 961 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 962 963config HARDLOCKUP_DETECTOR_PERF 964 bool 965 select SOFTLOCKUP_DETECTOR 966 967# 968# Enables a timestamp based low pass filter to compensate for perf based 969# hard lockup detection which runs too fast due to turbo modes. 970# 971config HARDLOCKUP_CHECK_TIMESTAMP 972 bool 973 974# 975# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard 976# lockup detector rather than the perf based detector. 977# 978config HARDLOCKUP_DETECTOR 979 bool "Detect Hard Lockups" 980 depends on DEBUG_KERNEL && !S390 981 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH 982 select LOCKUP_DETECTOR 983 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF 984 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH 985 help 986 Say Y here to enable the kernel to act as a watchdog to detect 987 hard lockups. 988 989 Hardlockups are bugs that cause the CPU to loop in kernel mode 990 for more than 10 seconds, without letting other interrupts have a 991 chance to run. The current stack trace is displayed upon detection 992 and the system will stay locked up. 993 994config BOOTPARAM_HARDLOCKUP_PANIC 995 bool "Panic (Reboot) On Hard Lockups" 996 depends on HARDLOCKUP_DETECTOR 997 help 998 Say Y here to enable the kernel to panic on "hard lockups", 999 which are bugs that cause the kernel to loop in kernel 1000 mode with interrupts disabled for more than 10 seconds (configurable 1001 using the watchdog_thresh sysctl). 1002 1003 Say N if unsure. 1004 1005config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 1006 int 1007 depends on HARDLOCKUP_DETECTOR 1008 range 0 1 1009 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 1010 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 1011 1012config DETECT_HUNG_TASK 1013 bool "Detect Hung Tasks" 1014 depends on DEBUG_KERNEL 1015 default SOFTLOCKUP_DETECTOR 1016 help 1017 Say Y here to enable the kernel to detect "hung tasks", 1018 which are bugs that cause the task to be stuck in 1019 uninterruptible "D" state indefinitely. 1020 1021 When a hung task is detected, the kernel will print the 1022 current stack trace (which you should report), but the 1023 task will stay in uninterruptible state. If lockdep is 1024 enabled then all held locks will also be reported. This 1025 feature has negligible overhead. 1026 1027config DEFAULT_HUNG_TASK_TIMEOUT 1028 int "Default timeout for hung task detection (in seconds)" 1029 depends on DETECT_HUNG_TASK 1030 default 120 1031 help 1032 This option controls the default timeout (in seconds) used 1033 to determine when a task has become non-responsive and should 1034 be considered hung. 1035 1036 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 1037 sysctl or by writing a value to 1038 /proc/sys/kernel/hung_task_timeout_secs. 1039 1040 A timeout of 0 disables the check. The default is two minutes. 1041 Keeping the default should be fine in most cases. 1042 1043config BOOTPARAM_HUNG_TASK_PANIC 1044 bool "Panic (Reboot) On Hung Tasks" 1045 depends on DETECT_HUNG_TASK 1046 help 1047 Say Y here to enable the kernel to panic on "hung tasks", 1048 which are bugs that cause the kernel to leave a task stuck 1049 in uninterruptible "D" state. 1050 1051 The panic can be used in combination with panic_timeout, 1052 to cause the system to reboot automatically after a 1053 hung task has been detected. This feature is useful for 1054 high-availability systems that have uptime guarantees and 1055 where a hung tasks must be resolved ASAP. 1056 1057 Say N if unsure. 1058 1059config BOOTPARAM_HUNG_TASK_PANIC_VALUE 1060 int 1061 depends on DETECT_HUNG_TASK 1062 range 0 1 1063 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 1064 default 1 if BOOTPARAM_HUNG_TASK_PANIC 1065 1066config WQ_WATCHDOG 1067 bool "Detect Workqueue Stalls" 1068 depends on DEBUG_KERNEL 1069 help 1070 Say Y here to enable stall detection on workqueues. If a 1071 worker pool doesn't make forward progress on a pending work 1072 item for over a given amount of time, 30s by default, a 1073 warning message is printed along with dump of workqueue 1074 state. This can be configured through kernel parameter 1075 "workqueue.watchdog_thresh" and its sysfs counterpart. 1076 1077config TEST_LOCKUP 1078 tristate "Test module to generate lockups" 1079 depends on m 1080 help 1081 This builds the "test_lockup" module that helps to make sure 1082 that watchdogs and lockup detectors are working properly. 1083 1084 Depending on module parameters it could emulate soft or hard 1085 lockup, "hung task", or locking arbitrary lock for a long time. 1086 Also it could generate series of lockups with cooling-down periods. 1087 1088 If unsure, say N. 1089 1090endmenu # "Debug lockups and hangs" 1091 1092menu "Scheduler Debugging" 1093 1094config SCHED_DEBUG 1095 bool "Collect scheduler debugging info" 1096 depends on DEBUG_KERNEL && PROC_FS 1097 default y 1098 help 1099 If you say Y here, the /proc/sched_debug file will be provided 1100 that can help debug the scheduler. The runtime overhead of this 1101 option is minimal. 1102 1103config SCHED_INFO 1104 bool 1105 default n 1106 1107config SCHEDSTATS 1108 bool "Collect scheduler statistics" 1109 depends on DEBUG_KERNEL && PROC_FS 1110 select SCHED_INFO 1111 help 1112 If you say Y here, additional code will be inserted into the 1113 scheduler and related routines to collect statistics about 1114 scheduler behavior and provide them in /proc/schedstat. These 1115 stats may be useful for both tuning and debugging the scheduler 1116 If you aren't debugging the scheduler or trying to tune a specific 1117 application, you can say N to avoid the very slight overhead 1118 this adds. 1119 1120endmenu 1121 1122config DEBUG_TIMEKEEPING 1123 bool "Enable extra timekeeping sanity checking" 1124 help 1125 This option will enable additional timekeeping sanity checks 1126 which may be helpful when diagnosing issues where timekeeping 1127 problems are suspected. 1128 1129 This may include checks in the timekeeping hotpaths, so this 1130 option may have a (very small) performance impact to some 1131 workloads. 1132 1133 If unsure, say N. 1134 1135config DEBUG_PREEMPT 1136 bool "Debug preemptible kernel" 1137 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT 1138 default y 1139 help 1140 If you say Y here then the kernel will use a debug variant of the 1141 commonly used smp_processor_id() function and will print warnings 1142 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1143 will detect preemption count underflows. 1144 1145menu "Lock Debugging (spinlocks, mutexes, etc...)" 1146 1147config LOCK_DEBUGGING_SUPPORT 1148 bool 1149 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1150 default y 1151 1152config PROVE_LOCKING 1153 bool "Lock debugging: prove locking correctness" 1154 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1155 select LOCKDEP 1156 select DEBUG_SPINLOCK 1157 select DEBUG_MUTEXES 1158 select DEBUG_RT_MUTEXES if RT_MUTEXES 1159 select DEBUG_RWSEMS 1160 select DEBUG_WW_MUTEX_SLOWPATH 1161 select DEBUG_LOCK_ALLOC 1162 select PREEMPT_COUNT if !ARCH_NO_PREEMPT 1163 select TRACE_IRQFLAGS 1164 default n 1165 help 1166 This feature enables the kernel to prove that all locking 1167 that occurs in the kernel runtime is mathematically 1168 correct: that under no circumstance could an arbitrary (and 1169 not yet triggered) combination of observed locking 1170 sequences (on an arbitrary number of CPUs, running an 1171 arbitrary number of tasks and interrupt contexts) cause a 1172 deadlock. 1173 1174 In short, this feature enables the kernel to report locking 1175 related deadlocks before they actually occur. 1176 1177 The proof does not depend on how hard and complex a 1178 deadlock scenario would be to trigger: how many 1179 participant CPUs, tasks and irq-contexts would be needed 1180 for it to trigger. The proof also does not depend on 1181 timing: if a race and a resulting deadlock is possible 1182 theoretically (no matter how unlikely the race scenario 1183 is), it will be proven so and will immediately be 1184 reported by the kernel (once the event is observed that 1185 makes the deadlock theoretically possible). 1186 1187 If a deadlock is impossible (i.e. the locking rules, as 1188 observed by the kernel, are mathematically correct), the 1189 kernel reports nothing. 1190 1191 NOTE: this feature can also be enabled for rwlocks, mutexes 1192 and rwsems - in which case all dependencies between these 1193 different locking variants are observed and mapped too, and 1194 the proof of observed correctness is also maintained for an 1195 arbitrary combination of these separate locking variants. 1196 1197 For more details, see Documentation/locking/lockdep-design.rst. 1198 1199config PROVE_RAW_LOCK_NESTING 1200 bool "Enable raw_spinlock - spinlock nesting checks" 1201 depends on PROVE_LOCKING 1202 default n 1203 help 1204 Enable the raw_spinlock vs. spinlock nesting checks which ensure 1205 that the lock nesting rules for PREEMPT_RT enabled kernels are 1206 not violated. 1207 1208 NOTE: There are known nesting problems. So if you enable this 1209 option expect lockdep splats until these problems have been fully 1210 addressed which is work in progress. This config switch allows to 1211 identify and analyze these problems. It will be removed and the 1212 check permanentely enabled once the main issues have been fixed. 1213 1214 If unsure, select N. 1215 1216config LOCK_STAT 1217 bool "Lock usage statistics" 1218 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1219 select LOCKDEP 1220 select DEBUG_SPINLOCK 1221 select DEBUG_MUTEXES 1222 select DEBUG_RT_MUTEXES if RT_MUTEXES 1223 select DEBUG_LOCK_ALLOC 1224 default n 1225 help 1226 This feature enables tracking lock contention points 1227 1228 For more details, see Documentation/locking/lockstat.rst 1229 1230 This also enables lock events required by "perf lock", 1231 subcommand of perf. 1232 If you want to use "perf lock", you also need to turn on 1233 CONFIG_EVENT_TRACING. 1234 1235 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1236 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1237 1238config DEBUG_RT_MUTEXES 1239 bool "RT Mutex debugging, deadlock detection" 1240 depends on DEBUG_KERNEL && RT_MUTEXES 1241 help 1242 This allows rt mutex semantics violations and rt mutex related 1243 deadlocks (lockups) to be detected and reported automatically. 1244 1245config DEBUG_SPINLOCK 1246 bool "Spinlock and rw-lock debugging: basic checks" 1247 depends on DEBUG_KERNEL 1248 select UNINLINE_SPIN_UNLOCK 1249 help 1250 Say Y here and build SMP to catch missing spinlock initialization 1251 and certain other kinds of spinlock errors commonly made. This is 1252 best used in conjunction with the NMI watchdog so that spinlock 1253 deadlocks are also debuggable. 1254 1255config DEBUG_MUTEXES 1256 bool "Mutex debugging: basic checks" 1257 depends on DEBUG_KERNEL 1258 help 1259 This feature allows mutex semantics violations to be detected and 1260 reported. 1261 1262config DEBUG_WW_MUTEX_SLOWPATH 1263 bool "Wait/wound mutex debugging: Slowpath testing" 1264 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1265 select DEBUG_LOCK_ALLOC 1266 select DEBUG_SPINLOCK 1267 select DEBUG_MUTEXES 1268 help 1269 This feature enables slowpath testing for w/w mutex users by 1270 injecting additional -EDEADLK wound/backoff cases. Together with 1271 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1272 will test all possible w/w mutex interface abuse with the 1273 exception of simply not acquiring all the required locks. 1274 Note that this feature can introduce significant overhead, so 1275 it really should not be enabled in a production or distro kernel, 1276 even a debug kernel. If you are a driver writer, enable it. If 1277 you are a distro, do not. 1278 1279config DEBUG_RWSEMS 1280 bool "RW Semaphore debugging: basic checks" 1281 depends on DEBUG_KERNEL 1282 help 1283 This debugging feature allows mismatched rw semaphore locks 1284 and unlocks to be detected and reported. 1285 1286config DEBUG_LOCK_ALLOC 1287 bool "Lock debugging: detect incorrect freeing of live locks" 1288 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1289 select DEBUG_SPINLOCK 1290 select DEBUG_MUTEXES 1291 select DEBUG_RT_MUTEXES if RT_MUTEXES 1292 select LOCKDEP 1293 help 1294 This feature will check whether any held lock (spinlock, rwlock, 1295 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1296 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1297 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1298 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1299 held during task exit. 1300 1301config LOCKDEP 1302 bool 1303 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1304 select STACKTRACE 1305 select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86 1306 select KALLSYMS 1307 select KALLSYMS_ALL 1308 1309config LOCKDEP_SMALL 1310 bool 1311 1312config DEBUG_LOCKDEP 1313 bool "Lock dependency engine debugging" 1314 depends on DEBUG_KERNEL && LOCKDEP 1315 help 1316 If you say Y here, the lock dependency engine will do 1317 additional runtime checks to debug itself, at the price 1318 of more runtime overhead. 1319 1320config DEBUG_ATOMIC_SLEEP 1321 bool "Sleep inside atomic section checking" 1322 select PREEMPT_COUNT 1323 depends on DEBUG_KERNEL 1324 depends on !ARCH_NO_PREEMPT 1325 help 1326 If you say Y here, various routines which may sleep will become very 1327 noisy if they are called inside atomic sections: when a spinlock is 1328 held, inside an rcu read side critical section, inside preempt disabled 1329 sections, inside an interrupt, etc... 1330 1331config DEBUG_LOCKING_API_SELFTESTS 1332 bool "Locking API boot-time self-tests" 1333 depends on DEBUG_KERNEL 1334 help 1335 Say Y here if you want the kernel to run a short self-test during 1336 bootup. The self-test checks whether common types of locking bugs 1337 are detected by debugging mechanisms or not. (if you disable 1338 lock debugging then those bugs wont be detected of course.) 1339 The following locking APIs are covered: spinlocks, rwlocks, 1340 mutexes and rwsems. 1341 1342config LOCK_TORTURE_TEST 1343 tristate "torture tests for locking" 1344 depends on DEBUG_KERNEL 1345 select TORTURE_TEST 1346 help 1347 This option provides a kernel module that runs torture tests 1348 on kernel locking primitives. The kernel module may be built 1349 after the fact on the running kernel to be tested, if desired. 1350 1351 Say Y here if you want kernel locking-primitive torture tests 1352 to be built into the kernel. 1353 Say M if you want these torture tests to build as a module. 1354 Say N if you are unsure. 1355 1356config WW_MUTEX_SELFTEST 1357 tristate "Wait/wound mutex selftests" 1358 help 1359 This option provides a kernel module that runs tests on the 1360 on the struct ww_mutex locking API. 1361 1362 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1363 with this test harness. 1364 1365 Say M if you want these self tests to build as a module. 1366 Say N if you are unsure. 1367 1368endmenu # lock debugging 1369 1370config TRACE_IRQFLAGS 1371 depends on TRACE_IRQFLAGS_SUPPORT 1372 bool 1373 help 1374 Enables hooks to interrupt enabling and disabling for 1375 either tracing or lock debugging. 1376 1377config TRACE_IRQFLAGS_NMI 1378 def_bool y 1379 depends on TRACE_IRQFLAGS 1380 depends on TRACE_IRQFLAGS_NMI_SUPPORT 1381 1382config STACKTRACE 1383 bool "Stack backtrace support" 1384 depends on STACKTRACE_SUPPORT 1385 help 1386 This option causes the kernel to create a /proc/pid/stack for 1387 every process, showing its current stack trace. 1388 It is also used by various kernel debugging features that require 1389 stack trace generation. 1390 1391config WARN_ALL_UNSEEDED_RANDOM 1392 bool "Warn for all uses of unseeded randomness" 1393 default n 1394 help 1395 Some parts of the kernel contain bugs relating to their use of 1396 cryptographically secure random numbers before it's actually possible 1397 to generate those numbers securely. This setting ensures that these 1398 flaws don't go unnoticed, by enabling a message, should this ever 1399 occur. This will allow people with obscure setups to know when things 1400 are going wrong, so that they might contact developers about fixing 1401 it. 1402 1403 Unfortunately, on some models of some architectures getting 1404 a fully seeded CRNG is extremely difficult, and so this can 1405 result in dmesg getting spammed for a surprisingly long 1406 time. This is really bad from a security perspective, and 1407 so architecture maintainers really need to do what they can 1408 to get the CRNG seeded sooner after the system is booted. 1409 However, since users cannot do anything actionable to 1410 address this, by default the kernel will issue only a single 1411 warning for the first use of unseeded randomness. 1412 1413 Say Y here if you want to receive warnings for all uses of 1414 unseeded randomness. This will be of use primarily for 1415 those developers interested in improving the security of 1416 Linux kernels running on their architecture (or 1417 subarchitecture). 1418 1419config DEBUG_KOBJECT 1420 bool "kobject debugging" 1421 depends on DEBUG_KERNEL 1422 help 1423 If you say Y here, some extra kobject debugging messages will be sent 1424 to the syslog. 1425 1426config DEBUG_KOBJECT_RELEASE 1427 bool "kobject release debugging" 1428 depends on DEBUG_OBJECTS_TIMERS 1429 help 1430 kobjects are reference counted objects. This means that their 1431 last reference count put is not predictable, and the kobject can 1432 live on past the point at which a driver decides to drop it's 1433 initial reference to the kobject gained on allocation. An 1434 example of this would be a struct device which has just been 1435 unregistered. 1436 1437 However, some buggy drivers assume that after such an operation, 1438 the memory backing the kobject can be immediately freed. This 1439 goes completely against the principles of a refcounted object. 1440 1441 If you say Y here, the kernel will delay the release of kobjects 1442 on the last reference count to improve the visibility of this 1443 kind of kobject release bug. 1444 1445config HAVE_DEBUG_BUGVERBOSE 1446 bool 1447 1448menu "Debug kernel data structures" 1449 1450config DEBUG_LIST 1451 bool "Debug linked list manipulation" 1452 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION 1453 help 1454 Enable this to turn on extended checks in the linked-list 1455 walking routines. 1456 1457 If unsure, say N. 1458 1459config DEBUG_PLIST 1460 bool "Debug priority linked list manipulation" 1461 depends on DEBUG_KERNEL 1462 help 1463 Enable this to turn on extended checks in the priority-ordered 1464 linked-list (plist) walking routines. This checks the entire 1465 list multiple times during each manipulation. 1466 1467 If unsure, say N. 1468 1469config DEBUG_SG 1470 bool "Debug SG table operations" 1471 depends on DEBUG_KERNEL 1472 help 1473 Enable this to turn on checks on scatter-gather tables. This can 1474 help find problems with drivers that do not properly initialize 1475 their sg tables. 1476 1477 If unsure, say N. 1478 1479config DEBUG_NOTIFIERS 1480 bool "Debug notifier call chains" 1481 depends on DEBUG_KERNEL 1482 help 1483 Enable this to turn on sanity checking for notifier call chains. 1484 This is most useful for kernel developers to make sure that 1485 modules properly unregister themselves from notifier chains. 1486 This is a relatively cheap check but if you care about maximum 1487 performance, say N. 1488 1489config BUG_ON_DATA_CORRUPTION 1490 bool "Trigger a BUG when data corruption is detected" 1491 select DEBUG_LIST 1492 help 1493 Select this option if the kernel should BUG when it encounters 1494 data corruption in kernel memory structures when they get checked 1495 for validity. 1496 1497 If unsure, say N. 1498 1499endmenu 1500 1501config DEBUG_CREDENTIALS 1502 bool "Debug credential management" 1503 depends on DEBUG_KERNEL 1504 help 1505 Enable this to turn on some debug checking for credential 1506 management. The additional code keeps track of the number of 1507 pointers from task_structs to any given cred struct, and checks to 1508 see that this number never exceeds the usage count of the cred 1509 struct. 1510 1511 Furthermore, if SELinux is enabled, this also checks that the 1512 security pointer in the cred struct is never seen to be invalid. 1513 1514 If unsure, say N. 1515 1516source "kernel/rcu/Kconfig.debug" 1517 1518config DEBUG_WQ_FORCE_RR_CPU 1519 bool "Force round-robin CPU selection for unbound work items" 1520 depends on DEBUG_KERNEL 1521 default n 1522 help 1523 Workqueue used to implicitly guarantee that work items queued 1524 without explicit CPU specified are put on the local CPU. This 1525 guarantee is no longer true and while local CPU is still 1526 preferred work items may be put on foreign CPUs. Kernel 1527 parameter "workqueue.debug_force_rr_cpu" is added to force 1528 round-robin CPU selection to flush out usages which depend on the 1529 now broken guarantee. This config option enables the debug 1530 feature by default. When enabled, memory and cache locality will 1531 be impacted. 1532 1533config DEBUG_BLOCK_EXT_DEVT 1534 bool "Force extended block device numbers and spread them" 1535 depends on DEBUG_KERNEL 1536 depends on BLOCK 1537 default n 1538 help 1539 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1540 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1541 YOU ARE DOING. Distros, please enable this and fix whatever 1542 is broken. 1543 1544 Conventionally, block device numbers are allocated from 1545 predetermined contiguous area. However, extended block area 1546 may introduce non-contiguous block device numbers. This 1547 option forces most block device numbers to be allocated from 1548 the extended space and spreads them to discover kernel or 1549 userland code paths which assume predetermined contiguous 1550 device number allocation. 1551 1552 Note that turning on this debug option shuffles all the 1553 device numbers for all IDE and SCSI devices including libata 1554 ones, so root partition specified using device number 1555 directly (via rdev or root=MAJ:MIN) won't work anymore. 1556 Textual device names (root=/dev/sdXn) will continue to work. 1557 1558 Say N if you are unsure. 1559 1560config CPU_HOTPLUG_STATE_CONTROL 1561 bool "Enable CPU hotplug state control" 1562 depends on DEBUG_KERNEL 1563 depends on HOTPLUG_CPU 1564 default n 1565 help 1566 Allows to write steps between "offline" and "online" to the CPUs 1567 sysfs target file so states can be stepped granular. This is a debug 1568 option for now as the hotplug machinery cannot be stopped and 1569 restarted at arbitrary points yet. 1570 1571 Say N if your are unsure. 1572 1573config LATENCYTOP 1574 bool "Latency measuring infrastructure" 1575 depends on DEBUG_KERNEL 1576 depends on STACKTRACE_SUPPORT 1577 depends on PROC_FS 1578 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1579 select KALLSYMS 1580 select KALLSYMS_ALL 1581 select STACKTRACE 1582 select SCHEDSTATS 1583 select SCHED_DEBUG 1584 help 1585 Enable this option if you want to use the LatencyTOP tool 1586 to find out which userspace is blocking on what kernel operations. 1587 1588source "kernel/trace/Kconfig" 1589 1590config PROVIDE_OHCI1394_DMA_INIT 1591 bool "Remote debugging over FireWire early on boot" 1592 depends on PCI && X86 1593 help 1594 If you want to debug problems which hang or crash the kernel early 1595 on boot and the crashing machine has a FireWire port, you can use 1596 this feature to remotely access the memory of the crashed machine 1597 over FireWire. This employs remote DMA as part of the OHCI1394 1598 specification which is now the standard for FireWire controllers. 1599 1600 With remote DMA, you can monitor the printk buffer remotely using 1601 firescope and access all memory below 4GB using fireproxy from gdb. 1602 Even controlling a kernel debugger is possible using remote DMA. 1603 1604 Usage: 1605 1606 If ohci1394_dma=early is used as boot parameter, it will initialize 1607 all OHCI1394 controllers which are found in the PCI config space. 1608 1609 As all changes to the FireWire bus such as enabling and disabling 1610 devices cause a bus reset and thereby disable remote DMA for all 1611 devices, be sure to have the cable plugged and FireWire enabled on 1612 the debugging host before booting the debug target for debugging. 1613 1614 This code (~1k) is freed after boot. By then, the firewire stack 1615 in charge of the OHCI-1394 controllers should be used instead. 1616 1617 See Documentation/core-api/debugging-via-ohci1394.rst for more information. 1618 1619source "samples/Kconfig" 1620 1621source "lib/Kconfig.kcsan" 1622 1623config ARCH_HAS_DEVMEM_IS_ALLOWED 1624 bool 1625 1626config STRICT_DEVMEM 1627 bool "Filter access to /dev/mem" 1628 depends on MMU && DEVMEM 1629 depends on ARCH_HAS_DEVMEM_IS_ALLOWED 1630 default y if PPC || X86 || ARM64 1631 help 1632 If this option is disabled, you allow userspace (root) access to all 1633 of memory, including kernel and userspace memory. Accidental 1634 access to this is obviously disastrous, but specific access can 1635 be used by people debugging the kernel. Note that with PAT support 1636 enabled, even in this case there are restrictions on /dev/mem 1637 use due to the cache aliasing requirements. 1638 1639 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem 1640 file only allows userspace access to PCI space and the BIOS code and 1641 data regions. This is sufficient for dosemu and X and all common 1642 users of /dev/mem. 1643 1644 If in doubt, say Y. 1645 1646config IO_STRICT_DEVMEM 1647 bool "Filter I/O access to /dev/mem" 1648 depends on STRICT_DEVMEM 1649 help 1650 If this option is disabled, you allow userspace (root) access to all 1651 io-memory regardless of whether a driver is actively using that 1652 range. Accidental access to this is obviously disastrous, but 1653 specific access can be used by people debugging kernel drivers. 1654 1655 If this option is switched on, the /dev/mem file only allows 1656 userspace access to *idle* io-memory ranges (see /proc/iomem) This 1657 may break traditional users of /dev/mem (dosemu, legacy X, etc...) 1658 if the driver using a given range cannot be disabled. 1659 1660 If in doubt, say Y. 1661 1662menu "$(SRCARCH) Debugging" 1663 1664source "arch/$(SRCARCH)/Kconfig.debug" 1665 1666endmenu 1667 1668menu "Kernel Testing and Coverage" 1669 1670source "lib/kunit/Kconfig" 1671 1672config NOTIFIER_ERROR_INJECTION 1673 tristate "Notifier error injection" 1674 depends on DEBUG_KERNEL 1675 select DEBUG_FS 1676 help 1677 This option provides the ability to inject artificial errors to 1678 specified notifier chain callbacks. It is useful to test the error 1679 handling of notifier call chain failures. 1680 1681 Say N if unsure. 1682 1683config PM_NOTIFIER_ERROR_INJECT 1684 tristate "PM notifier error injection module" 1685 depends on PM && NOTIFIER_ERROR_INJECTION 1686 default m if PM_DEBUG 1687 help 1688 This option provides the ability to inject artificial errors to 1689 PM notifier chain callbacks. It is controlled through debugfs 1690 interface /sys/kernel/debug/notifier-error-inject/pm 1691 1692 If the notifier call chain should be failed with some events 1693 notified, write the error code to "actions/<notifier event>/error". 1694 1695 Example: Inject PM suspend error (-12 = -ENOMEM) 1696 1697 # cd /sys/kernel/debug/notifier-error-inject/pm/ 1698 # echo -12 > actions/PM_SUSPEND_PREPARE/error 1699 # echo mem > /sys/power/state 1700 bash: echo: write error: Cannot allocate memory 1701 1702 To compile this code as a module, choose M here: the module will 1703 be called pm-notifier-error-inject. 1704 1705 If unsure, say N. 1706 1707config OF_RECONFIG_NOTIFIER_ERROR_INJECT 1708 tristate "OF reconfig notifier error injection module" 1709 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 1710 help 1711 This option provides the ability to inject artificial errors to 1712 OF reconfig notifier chain callbacks. It is controlled 1713 through debugfs interface under 1714 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 1715 1716 If the notifier call chain should be failed with some events 1717 notified, write the error code to "actions/<notifier event>/error". 1718 1719 To compile this code as a module, choose M here: the module will 1720 be called of-reconfig-notifier-error-inject. 1721 1722 If unsure, say N. 1723 1724config NETDEV_NOTIFIER_ERROR_INJECT 1725 tristate "Netdev notifier error injection module" 1726 depends on NET && NOTIFIER_ERROR_INJECTION 1727 help 1728 This option provides the ability to inject artificial errors to 1729 netdevice notifier chain callbacks. It is controlled through debugfs 1730 interface /sys/kernel/debug/notifier-error-inject/netdev 1731 1732 If the notifier call chain should be failed with some events 1733 notified, write the error code to "actions/<notifier event>/error". 1734 1735 Example: Inject netdevice mtu change error (-22 = -EINVAL) 1736 1737 # cd /sys/kernel/debug/notifier-error-inject/netdev 1738 # echo -22 > actions/NETDEV_CHANGEMTU/error 1739 # ip link set eth0 mtu 1024 1740 RTNETLINK answers: Invalid argument 1741 1742 To compile this code as a module, choose M here: the module will 1743 be called netdev-notifier-error-inject. 1744 1745 If unsure, say N. 1746 1747config FUNCTION_ERROR_INJECTION 1748 def_bool y 1749 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES 1750 1751config FAULT_INJECTION 1752 bool "Fault-injection framework" 1753 depends on DEBUG_KERNEL 1754 help 1755 Provide fault-injection framework. 1756 For more details, see Documentation/fault-injection/. 1757 1758config FAILSLAB 1759 bool "Fault-injection capability for kmalloc" 1760 depends on FAULT_INJECTION 1761 depends on SLAB || SLUB 1762 help 1763 Provide fault-injection capability for kmalloc. 1764 1765config FAIL_PAGE_ALLOC 1766 bool "Fault-injection capability for alloc_pages()" 1767 depends on FAULT_INJECTION 1768 help 1769 Provide fault-injection capability for alloc_pages(). 1770 1771config FAIL_MAKE_REQUEST 1772 bool "Fault-injection capability for disk IO" 1773 depends on FAULT_INJECTION && BLOCK 1774 help 1775 Provide fault-injection capability for disk IO. 1776 1777config FAIL_IO_TIMEOUT 1778 bool "Fault-injection capability for faking disk interrupts" 1779 depends on FAULT_INJECTION && BLOCK 1780 help 1781 Provide fault-injection capability on end IO handling. This 1782 will make the block layer "forget" an interrupt as configured, 1783 thus exercising the error handling. 1784 1785 Only works with drivers that use the generic timeout handling, 1786 for others it wont do anything. 1787 1788config FAIL_FUTEX 1789 bool "Fault-injection capability for futexes" 1790 select DEBUG_FS 1791 depends on FAULT_INJECTION && FUTEX 1792 help 1793 Provide fault-injection capability for futexes. 1794 1795config FAULT_INJECTION_DEBUG_FS 1796 bool "Debugfs entries for fault-injection capabilities" 1797 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1798 help 1799 Enable configuration of fault-injection capabilities via debugfs. 1800 1801config FAIL_FUNCTION 1802 bool "Fault-injection capability for functions" 1803 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION 1804 help 1805 Provide function-based fault-injection capability. 1806 This will allow you to override a specific function with a return 1807 with given return value. As a result, function caller will see 1808 an error value and have to handle it. This is useful to test the 1809 error handling in various subsystems. 1810 1811config FAIL_MMC_REQUEST 1812 bool "Fault-injection capability for MMC IO" 1813 depends on FAULT_INJECTION_DEBUG_FS && MMC 1814 help 1815 Provide fault-injection capability for MMC IO. 1816 This will make the mmc core return data errors. This is 1817 useful to test the error handling in the mmc block device 1818 and to test how the mmc host driver handles retries from 1819 the block device. 1820 1821config FAULT_INJECTION_STACKTRACE_FILTER 1822 bool "stacktrace filter for fault-injection capabilities" 1823 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1824 depends on !X86_64 1825 select STACKTRACE 1826 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1827 help 1828 Provide stacktrace filter for fault-injection capabilities 1829 1830config ARCH_HAS_KCOV 1831 bool 1832 help 1833 An architecture should select this when it can successfully 1834 build and run with CONFIG_KCOV. This typically requires 1835 disabling instrumentation for some early boot code. 1836 1837config CC_HAS_SANCOV_TRACE_PC 1838 def_bool $(cc-option,-fsanitize-coverage=trace-pc) 1839 1840 1841config KCOV 1842 bool "Code coverage for fuzzing" 1843 depends on ARCH_HAS_KCOV 1844 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS 1845 select DEBUG_FS 1846 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC 1847 help 1848 KCOV exposes kernel code coverage information in a form suitable 1849 for coverage-guided fuzzing (randomized testing). 1850 1851 If RANDOMIZE_BASE is enabled, PC values will not be stable across 1852 different machines and across reboots. If you need stable PC values, 1853 disable RANDOMIZE_BASE. 1854 1855 For more details, see Documentation/dev-tools/kcov.rst. 1856 1857config KCOV_ENABLE_COMPARISONS 1858 bool "Enable comparison operands collection by KCOV" 1859 depends on KCOV 1860 depends on $(cc-option,-fsanitize-coverage=trace-cmp) 1861 help 1862 KCOV also exposes operands of every comparison in the instrumented 1863 code along with operand sizes and PCs of the comparison instructions. 1864 These operands can be used by fuzzing engines to improve the quality 1865 of fuzzing coverage. 1866 1867config KCOV_INSTRUMENT_ALL 1868 bool "Instrument all code by default" 1869 depends on KCOV 1870 default y 1871 help 1872 If you are doing generic system call fuzzing (like e.g. syzkaller), 1873 then you will want to instrument the whole kernel and you should 1874 say y here. If you are doing more targeted fuzzing (like e.g. 1875 filesystem fuzzing with AFL) then you will want to enable coverage 1876 for more specific subsets of files, and should say n here. 1877 1878config KCOV_IRQ_AREA_SIZE 1879 hex "Size of interrupt coverage collection area in words" 1880 depends on KCOV 1881 default 0x40000 1882 help 1883 KCOV uses preallocated per-cpu areas to collect coverage from 1884 soft interrupts. This specifies the size of those areas in the 1885 number of unsigned long words. 1886 1887menuconfig RUNTIME_TESTING_MENU 1888 bool "Runtime Testing" 1889 def_bool y 1890 1891if RUNTIME_TESTING_MENU 1892 1893config LKDTM 1894 tristate "Linux Kernel Dump Test Tool Module" 1895 depends on DEBUG_FS 1896 help 1897 This module enables testing of the different dumping mechanisms by 1898 inducing system failures at predefined crash points. 1899 If you don't need it: say N 1900 Choose M here to compile this code as a module. The module will be 1901 called lkdtm. 1902 1903 Documentation on how to use the module can be found in 1904 Documentation/fault-injection/provoke-crashes.rst 1905 1906config TEST_LIST_SORT 1907 tristate "Linked list sorting test" 1908 depends on DEBUG_KERNEL || m 1909 help 1910 Enable this to turn on 'list_sort()' function test. This test is 1911 executed only once during system boot (so affects only boot time), 1912 or at module load time. 1913 1914 If unsure, say N. 1915 1916config TEST_MIN_HEAP 1917 tristate "Min heap test" 1918 depends on DEBUG_KERNEL || m 1919 help 1920 Enable this to turn on min heap function tests. This test is 1921 executed only once during system boot (so affects only boot time), 1922 or at module load time. 1923 1924 If unsure, say N. 1925 1926config TEST_SORT 1927 tristate "Array-based sort test" 1928 depends on DEBUG_KERNEL || m 1929 help 1930 This option enables the self-test function of 'sort()' at boot, 1931 or at module load time. 1932 1933 If unsure, say N. 1934 1935config KPROBES_SANITY_TEST 1936 bool "Kprobes sanity tests" 1937 depends on DEBUG_KERNEL 1938 depends on KPROBES 1939 help 1940 This option provides for testing basic kprobes functionality on 1941 boot. Samples of kprobe and kretprobe are inserted and 1942 verified for functionality. 1943 1944 Say N if you are unsure. 1945 1946config BACKTRACE_SELF_TEST 1947 tristate "Self test for the backtrace code" 1948 depends on DEBUG_KERNEL 1949 help 1950 This option provides a kernel module that can be used to test 1951 the kernel stack backtrace code. This option is not useful 1952 for distributions or general kernels, but only for kernel 1953 developers working on architecture code. 1954 1955 Note that if you want to also test saved backtraces, you will 1956 have to enable STACKTRACE as well. 1957 1958 Say N if you are unsure. 1959 1960config RBTREE_TEST 1961 tristate "Red-Black tree test" 1962 depends on DEBUG_KERNEL 1963 help 1964 A benchmark measuring the performance of the rbtree library. 1965 Also includes rbtree invariant checks. 1966 1967config REED_SOLOMON_TEST 1968 tristate "Reed-Solomon library test" 1969 depends on DEBUG_KERNEL || m 1970 select REED_SOLOMON 1971 select REED_SOLOMON_ENC16 1972 select REED_SOLOMON_DEC16 1973 help 1974 This option enables the self-test function of rslib at boot, 1975 or at module load time. 1976 1977 If unsure, say N. 1978 1979config INTERVAL_TREE_TEST 1980 tristate "Interval tree test" 1981 depends on DEBUG_KERNEL 1982 select INTERVAL_TREE 1983 help 1984 A benchmark measuring the performance of the interval tree library 1985 1986config PERCPU_TEST 1987 tristate "Per cpu operations test" 1988 depends on m && DEBUG_KERNEL 1989 help 1990 Enable this option to build test module which validates per-cpu 1991 operations. 1992 1993 If unsure, say N. 1994 1995config ATOMIC64_SELFTEST 1996 tristate "Perform an atomic64_t self-test" 1997 help 1998 Enable this option to test the atomic64_t functions at boot or 1999 at module load time. 2000 2001 If unsure, say N. 2002 2003config ASYNC_RAID6_TEST 2004 tristate "Self test for hardware accelerated raid6 recovery" 2005 depends on ASYNC_RAID6_RECOV 2006 select ASYNC_MEMCPY 2007 help 2008 This is a one-shot self test that permutes through the 2009 recovery of all the possible two disk failure scenarios for a 2010 N-disk array. Recovery is performed with the asynchronous 2011 raid6 recovery routines, and will optionally use an offload 2012 engine if one is available. 2013 2014 If unsure, say N. 2015 2016config TEST_HEXDUMP 2017 tristate "Test functions located in the hexdump module at runtime" 2018 2019config TEST_STRING_HELPERS 2020 tristate "Test functions located in the string_helpers module at runtime" 2021 2022config TEST_STRSCPY 2023 tristate "Test strscpy*() family of functions at runtime" 2024 2025config TEST_KSTRTOX 2026 tristate "Test kstrto*() family of functions at runtime" 2027 2028config TEST_PRINTF 2029 tristate "Test printf() family of functions at runtime" 2030 2031config TEST_BITMAP 2032 tristate "Test bitmap_*() family of functions at runtime" 2033 help 2034 Enable this option to test the bitmap functions at boot. 2035 2036 If unsure, say N. 2037 2038config TEST_BITFIELD 2039 tristate "Test bitfield functions at runtime" 2040 help 2041 Enable this option to test the bitfield functions at boot. 2042 2043 If unsure, say N. 2044 2045config TEST_UUID 2046 tristate "Test functions located in the uuid module at runtime" 2047 2048config TEST_XARRAY 2049 tristate "Test the XArray code at runtime" 2050 2051config TEST_OVERFLOW 2052 tristate "Test check_*_overflow() functions at runtime" 2053 2054config TEST_RHASHTABLE 2055 tristate "Perform selftest on resizable hash table" 2056 help 2057 Enable this option to test the rhashtable functions at boot. 2058 2059 If unsure, say N. 2060 2061config TEST_HASH 2062 tristate "Perform selftest on hash functions" 2063 help 2064 Enable this option to test the kernel's integer (<linux/hash.h>), 2065 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>) 2066 hash functions on boot (or module load). 2067 2068 This is intended to help people writing architecture-specific 2069 optimized versions. If unsure, say N. 2070 2071config TEST_IDA 2072 tristate "Perform selftest on IDA functions" 2073 2074config TEST_PARMAN 2075 tristate "Perform selftest on priority array manager" 2076 depends on PARMAN 2077 help 2078 Enable this option to test priority array manager on boot 2079 (or module load). 2080 2081 If unsure, say N. 2082 2083config TEST_IRQ_TIMINGS 2084 bool "IRQ timings selftest" 2085 depends on IRQ_TIMINGS 2086 help 2087 Enable this option to test the irq timings code on boot. 2088 2089 If unsure, say N. 2090 2091config TEST_LKM 2092 tristate "Test module loading with 'hello world' module" 2093 depends on m 2094 help 2095 This builds the "test_module" module that emits "Hello, world" 2096 on printk when loaded. It is designed to be used for basic 2097 evaluation of the module loading subsystem (for example when 2098 validating module verification). It lacks any extra dependencies, 2099 and will not normally be loaded by the system unless explicitly 2100 requested by name. 2101 2102 If unsure, say N. 2103 2104config TEST_BITOPS 2105 tristate "Test module for compilation of bitops operations" 2106 depends on m 2107 help 2108 This builds the "test_bitops" module that is much like the 2109 TEST_LKM module except that it does a basic exercise of the 2110 set/clear_bit macros and get_count_order/long to make sure there are 2111 no compiler warnings from C=1 sparse checker or -Wextra 2112 compilations. It has no dependencies and doesn't run or load unless 2113 explicitly requested by name. for example: modprobe test_bitops. 2114 2115 If unsure, say N. 2116 2117config TEST_VMALLOC 2118 tristate "Test module for stress/performance analysis of vmalloc allocator" 2119 default n 2120 depends on MMU 2121 depends on m 2122 help 2123 This builds the "test_vmalloc" module that should be used for 2124 stress and performance analysis. So, any new change for vmalloc 2125 subsystem can be evaluated from performance and stability point 2126 of view. 2127 2128 If unsure, say N. 2129 2130config TEST_USER_COPY 2131 tristate "Test user/kernel boundary protections" 2132 depends on m 2133 help 2134 This builds the "test_user_copy" module that runs sanity checks 2135 on the copy_to/from_user infrastructure, making sure basic 2136 user/kernel boundary testing is working. If it fails to load, 2137 a regression has been detected in the user/kernel memory boundary 2138 protections. 2139 2140 If unsure, say N. 2141 2142config TEST_BPF 2143 tristate "Test BPF filter functionality" 2144 depends on m && NET 2145 help 2146 This builds the "test_bpf" module that runs various test vectors 2147 against the BPF interpreter or BPF JIT compiler depending on the 2148 current setting. This is in particular useful for BPF JIT compiler 2149 development, but also to run regression tests against changes in 2150 the interpreter code. It also enables test stubs for eBPF maps and 2151 verifier used by user space verifier testsuite. 2152 2153 If unsure, say N. 2154 2155config TEST_BLACKHOLE_DEV 2156 tristate "Test blackhole netdev functionality" 2157 depends on m && NET 2158 help 2159 This builds the "test_blackhole_dev" module that validates the 2160 data path through this blackhole netdev. 2161 2162 If unsure, say N. 2163 2164config FIND_BIT_BENCHMARK 2165 tristate "Test find_bit functions" 2166 help 2167 This builds the "test_find_bit" module that measure find_*_bit() 2168 functions performance. 2169 2170 If unsure, say N. 2171 2172config TEST_FIRMWARE 2173 tristate "Test firmware loading via userspace interface" 2174 depends on FW_LOADER 2175 help 2176 This builds the "test_firmware" module that creates a userspace 2177 interface for testing firmware loading. This can be used to 2178 control the triggering of firmware loading without needing an 2179 actual firmware-using device. The contents can be rechecked by 2180 userspace. 2181 2182 If unsure, say N. 2183 2184config TEST_SYSCTL 2185 tristate "sysctl test driver" 2186 depends on PROC_SYSCTL 2187 help 2188 This builds the "test_sysctl" module. This driver enables to test the 2189 proc sysctl interfaces available to drivers safely without affecting 2190 production knobs which might alter system functionality. 2191 2192 If unsure, say N. 2193 2194config SYSCTL_KUNIT_TEST 2195 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS 2196 depends on KUNIT 2197 default KUNIT_ALL_TESTS 2198 help 2199 This builds the proc sysctl unit test, which runs on boot. 2200 Tests the API contract and implementation correctness of sysctl. 2201 For more information on KUnit and unit tests in general please refer 2202 to the KUnit documentation in Documentation/dev-tools/kunit/. 2203 2204 If unsure, say N. 2205 2206config LIST_KUNIT_TEST 2207 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS 2208 depends on KUNIT 2209 default KUNIT_ALL_TESTS 2210 help 2211 This builds the linked list KUnit test suite. 2212 It tests that the API and basic functionality of the list_head type 2213 and associated macros. 2214 2215 KUnit tests run during boot and output the results to the debug log 2216 in TAP format (https://testanything.org/). Only useful for kernel devs 2217 running the KUnit test harness, and not intended for inclusion into a 2218 production build. 2219 2220 For more information on KUnit and unit tests in general please refer 2221 to the KUnit documentation in Documentation/dev-tools/kunit/. 2222 2223 If unsure, say N. 2224 2225config LINEAR_RANGES_TEST 2226 tristate "KUnit test for linear_ranges" 2227 depends on KUNIT 2228 select LINEAR_RANGES 2229 help 2230 This builds the linear_ranges unit test, which runs on boot. 2231 Tests the linear_ranges logic correctness. 2232 For more information on KUnit and unit tests in general please refer 2233 to the KUnit documentation in Documentation/dev-tools/kunit/. 2234 2235 If unsure, say N. 2236 2237config BITS_TEST 2238 tristate "KUnit test for bits.h" 2239 depends on KUNIT 2240 help 2241 This builds the bits unit test. 2242 Tests the logic of macros defined in bits.h. 2243 For more information on KUnit and unit tests in general please refer 2244 to the KUnit documentation in Documentation/dev-tools/kunit/. 2245 2246 If unsure, say N. 2247 2248config TEST_UDELAY 2249 tristate "udelay test driver" 2250 help 2251 This builds the "udelay_test" module that helps to make sure 2252 that udelay() is working properly. 2253 2254 If unsure, say N. 2255 2256config TEST_STATIC_KEYS 2257 tristate "Test static keys" 2258 depends on m 2259 help 2260 Test the static key interfaces. 2261 2262 If unsure, say N. 2263 2264config TEST_KMOD 2265 tristate "kmod stress tester" 2266 depends on m 2267 depends on NETDEVICES && NET_CORE && INET # for TUN 2268 depends on BLOCK 2269 select TEST_LKM 2270 select XFS_FS 2271 select TUN 2272 select BTRFS_FS 2273 help 2274 Test the kernel's module loading mechanism: kmod. kmod implements 2275 support to load modules using the Linux kernel's usermode helper. 2276 This test provides a series of tests against kmod. 2277 2278 Although technically you can either build test_kmod as a module or 2279 into the kernel we disallow building it into the kernel since 2280 it stress tests request_module() and this will very likely cause 2281 some issues by taking over precious threads available from other 2282 module load requests, ultimately this could be fatal. 2283 2284 To run tests run: 2285 2286 tools/testing/selftests/kmod/kmod.sh --help 2287 2288 If unsure, say N. 2289 2290config TEST_DEBUG_VIRTUAL 2291 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 2292 depends on DEBUG_VIRTUAL 2293 help 2294 Test the kernel's ability to detect incorrect calls to 2295 virt_to_phys() done against the non-linear part of the 2296 kernel's virtual address map. 2297 2298 If unsure, say N. 2299 2300config TEST_MEMCAT_P 2301 tristate "Test memcat_p() helper function" 2302 help 2303 Test the memcat_p() helper for correctly merging two 2304 pointer arrays together. 2305 2306 If unsure, say N. 2307 2308config TEST_LIVEPATCH 2309 tristate "Test livepatching" 2310 default n 2311 depends on DYNAMIC_DEBUG 2312 depends on LIVEPATCH 2313 depends on m 2314 help 2315 Test kernel livepatching features for correctness. The tests will 2316 load test modules that will be livepatched in various scenarios. 2317 2318 To run all the livepatching tests: 2319 2320 make -C tools/testing/selftests TARGETS=livepatch run_tests 2321 2322 Alternatively, individual tests may be invoked: 2323 2324 tools/testing/selftests/livepatch/test-callbacks.sh 2325 tools/testing/selftests/livepatch/test-livepatch.sh 2326 tools/testing/selftests/livepatch/test-shadow-vars.sh 2327 2328 If unsure, say N. 2329 2330config TEST_OBJAGG 2331 tristate "Perform selftest on object aggreration manager" 2332 default n 2333 depends on OBJAGG 2334 help 2335 Enable this option to test object aggregation manager on boot 2336 (or module load). 2337 2338 2339config TEST_STACKINIT 2340 tristate "Test level of stack variable initialization" 2341 help 2342 Test if the kernel is zero-initializing stack variables and 2343 padding. Coverage is controlled by compiler flags, 2344 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF, 2345 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. 2346 2347 If unsure, say N. 2348 2349config TEST_MEMINIT 2350 tristate "Test heap/page initialization" 2351 help 2352 Test if the kernel is zero-initializing heap and page allocations. 2353 This can be useful to test init_on_alloc and init_on_free features. 2354 2355 If unsure, say N. 2356 2357config TEST_HMM 2358 tristate "Test HMM (Heterogeneous Memory Management)" 2359 depends on TRANSPARENT_HUGEPAGE 2360 depends on DEVICE_PRIVATE 2361 select HMM_MIRROR 2362 select MMU_NOTIFIER 2363 help 2364 This is a pseudo device driver solely for testing HMM. 2365 Say M here if you want to build the HMM test module. 2366 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. 2367 2368 If unsure, say N. 2369 2370config TEST_FPU 2371 tristate "Test floating point operations in kernel space" 2372 depends on X86 && !KCOV_INSTRUMENT_ALL 2373 help 2374 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu 2375 which will trigger a sequence of floating point operations. This is used 2376 for self-testing floating point control register setting in 2377 kernel_fpu_begin(). 2378 2379 If unsure, say N. 2380 2381endif # RUNTIME_TESTING_MENU 2382 2383config MEMTEST 2384 bool "Memtest" 2385 help 2386 This option adds a kernel parameter 'memtest', which allows memtest 2387 to be set. 2388 memtest=0, mean disabled; -- default 2389 memtest=1, mean do 1 test pattern; 2390 ... 2391 memtest=17, mean do 17 test patterns. 2392 If you are unsure how to answer this question, answer N. 2393 2394 2395 2396config HYPERV_TESTING 2397 bool "Microsoft Hyper-V driver testing" 2398 default n 2399 depends on HYPERV && DEBUG_FS 2400 help 2401 Select this option to enable Hyper-V vmbus testing. 2402 2403endmenu # "Kernel Testing and Coverage" 2404 2405endmenu # Kernel hacking 2406