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 215config DEBUG_INFO_REDUCED 216 bool "Reduce debugging information" 217 depends on DEBUG_INFO 218 help 219 If you say Y here gcc is instructed to generate less debugging 220 information for structure types. This means that tools that 221 need full debugging information (like kgdb or systemtap) won't 222 be happy. But if you merely need debugging information to 223 resolve line numbers there is no loss. Advantage is that 224 build directory object sizes shrink dramatically over a full 225 DEBUG_INFO build and compile times are reduced too. 226 Only works with newer gcc versions. 227 228config DEBUG_INFO_COMPRESSED 229 bool "Compressed debugging information" 230 depends on DEBUG_INFO 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 DEBUG_INFO 247 depends on $(cc-option,-gsplit-dwarf) 248 help 249 Generate debug info into separate .dwo files. This significantly 250 reduces the build directory size for builds with DEBUG_INFO, 251 because it stores the information only once on disk in .dwo 252 files instead of multiple times in object files and executables. 253 In addition the debug information is also compressed. 254 255 Requires recent gcc (4.7+) and recent gdb/binutils. 256 Any tool that packages or reads debug information would need 257 to know about the .dwo files and include them. 258 Incompatible with older versions of ccache. 259 260config DEBUG_INFO_DWARF4 261 bool "Generate dwarf4 debuginfo" 262 depends on DEBUG_INFO 263 depends on $(cc-option,-gdwarf-4) 264 help 265 Generate dwarf4 debug info. This requires recent versions 266 of gcc and gdb. It makes the debug information larger. 267 But it significantly improves the success of resolving 268 variables in gdb on optimized code. 269 270config DEBUG_INFO_BTF 271 bool "Generate BTF typeinfo" 272 depends on DEBUG_INFO 273 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED 274 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST 275 help 276 Generate deduplicated BTF type information from DWARF debug info. 277 Turning this on expects presence of pahole tool, which will convert 278 DWARF type info into equivalent deduplicated BTF type info. 279 280config GDB_SCRIPTS 281 bool "Provide GDB scripts for kernel debugging" 282 depends on DEBUG_INFO 283 help 284 This creates the required links to GDB helper scripts in the 285 build directory. If you load vmlinux into gdb, the helper 286 scripts will be automatically imported by gdb as well, and 287 additional functions are available to analyze a Linux kernel 288 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst 289 for further details. 290 291config ENABLE_MUST_CHECK 292 bool "Enable __must_check logic" 293 default y 294 help 295 Enable the __must_check logic in the kernel build. Disable this to 296 suppress the "warning: ignoring return value of 'foo', declared with 297 attribute warn_unused_result" messages. 298 299config FRAME_WARN 300 int "Warn for stack frames larger than" 301 range 0 8192 302 default 2048 if GCC_PLUGIN_LATENT_ENTROPY 303 default 1280 if (!64BIT && PARISC) 304 default 1024 if (!64BIT && !PARISC) 305 default 2048 if 64BIT 306 help 307 Tell gcc to warn at build time for stack frames larger than this. 308 Setting this too low will cause a lot of warnings. 309 Setting it to 0 disables the warning. 310 311config STRIP_ASM_SYMS 312 bool "Strip assembler-generated symbols during link" 313 default n 314 help 315 Strip internal assembler-generated symbols during a link (symbols 316 that look like '.Lxxx') so they don't pollute the output of 317 get_wchan() and suchlike. 318 319config READABLE_ASM 320 bool "Generate readable assembler code" 321 depends on DEBUG_KERNEL 322 help 323 Disable some compiler optimizations that tend to generate human unreadable 324 assembler output. This may make the kernel slightly slower, but it helps 325 to keep kernel developers who have to stare a lot at assembler listings 326 sane. 327 328config HEADERS_INSTALL 329 bool "Install uapi headers to usr/include" 330 depends on !UML 331 help 332 This option will install uapi headers (headers exported to user-space) 333 into the usr/include directory for use during the kernel build. 334 This is unneeded for building the kernel itself, but needed for some 335 user-space program samples. It is also needed by some features such 336 as uapi header sanity checks. 337 338config DEBUG_SECTION_MISMATCH 339 bool "Enable full Section mismatch analysis" 340 help 341 The section mismatch analysis checks if there are illegal 342 references from one section to another section. 343 During linktime or runtime, some sections are dropped; 344 any use of code/data previously in these sections would 345 most likely result in an oops. 346 In the code, functions and variables are annotated with 347 __init,, etc. (see the full list in include/linux/init.h), 348 which results in the code/data being placed in specific sections. 349 The section mismatch analysis is always performed after a full 350 kernel build, and enabling this option causes the following 351 additional step to occur: 352 - Add the option -fno-inline-functions-called-once to gcc commands. 353 When inlining a function annotated with __init in a non-init 354 function, we would lose the section information and thus 355 the analysis would not catch the illegal reference. 356 This option tells gcc to inline less (but it does result in 357 a larger kernel). 358 359config SECTION_MISMATCH_WARN_ONLY 360 bool "Make section mismatch errors non-fatal" 361 default y 362 help 363 If you say N here, the build process will fail if there are any 364 section mismatch, instead of just throwing warnings. 365 366 If unsure, say Y. 367 368config DEBUG_FORCE_FUNCTION_ALIGN_32B 369 bool "Force all function address 32B aligned" if EXPERT 370 help 371 There are cases that a commit from one domain changes the function 372 address alignment of other domains, and cause magic performance 373 bump (regression or improvement). Enable this option will help to 374 verify if the bump is caused by function alignment changes, while 375 it will slightly increase the kernel size and affect icache usage. 376 377 It is mainly for debug and performance tuning use. 378 379# 380# Select this config option from the architecture Kconfig, if it 381# is preferred to always offer frame pointers as a config 382# option on the architecture (regardless of KERNEL_DEBUG): 383# 384config ARCH_WANT_FRAME_POINTERS 385 bool 386 387config FRAME_POINTER 388 bool "Compile the kernel with frame pointers" 389 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS 390 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 391 help 392 If you say Y here the resulting kernel image will be slightly 393 larger and slower, but it gives very useful debugging information 394 in case of kernel bugs. (precise oopses/stacktraces/warnings) 395 396config STACK_VALIDATION 397 bool "Compile-time stack metadata validation" 398 depends on HAVE_STACK_VALIDATION 399 default n 400 help 401 Add compile-time checks to validate stack metadata, including frame 402 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure 403 that runtime stack traces are more reliable. 404 405 This is also a prerequisite for generation of ORC unwind data, which 406 is needed for CONFIG_UNWINDER_ORC. 407 408 For more information, see 409 tools/objtool/Documentation/stack-validation.txt. 410 411config VMLINUX_VALIDATION 412 bool 413 depends on STACK_VALIDATION && DEBUG_ENTRY && !PARAVIRT 414 default y 415 416config DEBUG_FORCE_WEAK_PER_CPU 417 bool "Force weak per-cpu definitions" 418 depends on DEBUG_KERNEL 419 help 420 s390 and alpha require percpu variables in modules to be 421 defined weak to work around addressing range issue which 422 puts the following two restrictions on percpu variable 423 definitions. 424 425 1. percpu symbols must be unique whether static or not 426 2. percpu variables can't be defined inside a function 427 428 To ensure that generic code follows the above rules, this 429 option forces all percpu variables to be defined as weak. 430 431endmenu # "Compiler options" 432 433menu "Generic Kernel Debugging Instruments" 434 435config MAGIC_SYSRQ 436 bool "Magic SysRq key" 437 depends on !UML 438 help 439 If you say Y here, you will have some control over the system even 440 if the system crashes for example during kernel debugging (e.g., you 441 will be able to flush the buffer cache to disk, reboot the system 442 immediately or dump some status information). This is accomplished 443 by pressing various keys while holding SysRq (Alt+PrintScreen). It 444 also works on a serial console (on PC hardware at least), if you 445 send a BREAK and then within 5 seconds a command keypress. The 446 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 447 Don't say Y unless you really know what this hack does. 448 449config MAGIC_SYSRQ_DEFAULT_ENABLE 450 hex "Enable magic SysRq key functions by default" 451 depends on MAGIC_SYSRQ 452 default 0x1 453 help 454 Specifies which SysRq key functions are enabled by default. 455 This may be set to 1 or 0 to enable or disable them all, or 456 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 457 458config MAGIC_SYSRQ_SERIAL 459 bool "Enable magic SysRq key over serial" 460 depends on MAGIC_SYSRQ 461 default y 462 help 463 Many embedded boards have a disconnected TTL level serial which can 464 generate some garbage that can lead to spurious false sysrq detects. 465 This option allows you to decide whether you want to enable the 466 magic SysRq key. 467 468config MAGIC_SYSRQ_SERIAL_SEQUENCE 469 string "Char sequence that enables magic SysRq over serial" 470 depends on MAGIC_SYSRQ_SERIAL 471 default "" 472 help 473 Specifies a sequence of characters that can follow BREAK to enable 474 SysRq on a serial console. 475 476 If unsure, leave an empty string and the option will not be enabled. 477 478config DEBUG_FS 479 bool "Debug Filesystem" 480 help 481 debugfs is a virtual file system that kernel developers use to put 482 debugging files into. Enable this option to be able to read and 483 write to these files. 484 485 For detailed documentation on the debugfs API, see 486 Documentation/filesystems/. 487 488 If unsure, say N. 489 490choice 491 prompt "Debugfs default access" 492 depends on DEBUG_FS 493 default DEBUG_FS_ALLOW_ALL 494 help 495 This selects the default access restrictions for debugfs. 496 It can be overridden with kernel command line option 497 debugfs=[on,no-mount,off]. The restrictions apply for API access 498 and filesystem registration. 499 500config DEBUG_FS_ALLOW_ALL 501 bool "Access normal" 502 help 503 No restrictions apply. Both API and filesystem registration 504 is on. This is the normal default operation. 505 506config DEBUG_FS_DISALLOW_MOUNT 507 bool "Do not register debugfs as filesystem" 508 help 509 The API is open but filesystem is not loaded. Clients can still do 510 their work and read with debug tools that do not need 511 debugfs filesystem. 512 513config DEBUG_FS_ALLOW_NONE 514 bool "No access" 515 help 516 Access is off. Clients get -PERM when trying to create nodes in 517 debugfs tree and debugfs is not registered as a filesystem. 518 Client can then back-off or continue without debugfs access. 519 520endchoice 521 522source "lib/Kconfig.kgdb" 523source "lib/Kconfig.ubsan" 524source "lib/Kconfig.kcsan" 525 526endmenu 527 528config DEBUG_KERNEL 529 bool "Kernel debugging" 530 help 531 Say Y here if you are developing drivers or trying to debug and 532 identify kernel problems. 533 534config DEBUG_MISC 535 bool "Miscellaneous debug code" 536 default DEBUG_KERNEL 537 depends on DEBUG_KERNEL 538 help 539 Say Y here if you need to enable miscellaneous debug code that should 540 be under a more specific debug option but isn't. 541 542 543menu "Memory Debugging" 544 545source "mm/Kconfig.debug" 546 547config DEBUG_OBJECTS 548 bool "Debug object operations" 549 depends on DEBUG_KERNEL 550 help 551 If you say Y here, additional code will be inserted into the 552 kernel to track the life time of various objects and validate 553 the operations on those objects. 554 555config DEBUG_OBJECTS_SELFTEST 556 bool "Debug objects selftest" 557 depends on DEBUG_OBJECTS 558 help 559 This enables the selftest of the object debug code. 560 561config DEBUG_OBJECTS_FREE 562 bool "Debug objects in freed memory" 563 depends on DEBUG_OBJECTS 564 help 565 This enables checks whether a k/v free operation frees an area 566 which contains an object which has not been deactivated 567 properly. This can make kmalloc/kfree-intensive workloads 568 much slower. 569 570config DEBUG_OBJECTS_TIMERS 571 bool "Debug timer objects" 572 depends on DEBUG_OBJECTS 573 help 574 If you say Y here, additional code will be inserted into the 575 timer routines to track the life time of timer objects and 576 validate the timer operations. 577 578config DEBUG_OBJECTS_WORK 579 bool "Debug work objects" 580 depends on DEBUG_OBJECTS 581 help 582 If you say Y here, additional code will be inserted into the 583 work queue routines to track the life time of work objects and 584 validate the work operations. 585 586config DEBUG_OBJECTS_RCU_HEAD 587 bool "Debug RCU callbacks objects" 588 depends on DEBUG_OBJECTS 589 help 590 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 591 592config DEBUG_OBJECTS_PERCPU_COUNTER 593 bool "Debug percpu counter objects" 594 depends on DEBUG_OBJECTS 595 help 596 If you say Y here, additional code will be inserted into the 597 percpu counter routines to track the life time of percpu counter 598 objects and validate the percpu counter operations. 599 600config DEBUG_OBJECTS_ENABLE_DEFAULT 601 int "debug_objects bootup default value (0-1)" 602 range 0 1 603 default "1" 604 depends on DEBUG_OBJECTS 605 help 606 Debug objects boot parameter default value 607 608config DEBUG_SLAB 609 bool "Debug slab memory allocations" 610 depends on DEBUG_KERNEL && SLAB 611 help 612 Say Y here to have the kernel do limited verification on memory 613 allocation as well as poisoning memory on free to catch use of freed 614 memory. This can make kmalloc/kfree-intensive workloads much slower. 615 616config SLUB_DEBUG_ON 617 bool "SLUB debugging on by default" 618 depends on SLUB && SLUB_DEBUG 619 default n 620 help 621 Boot with debugging on by default. SLUB boots by default with 622 the runtime debug capabilities switched off. Enabling this is 623 equivalent to specifying the "slub_debug" parameter on boot. 624 There is no support for more fine grained debug control like 625 possible with slub_debug=xxx. SLUB debugging may be switched 626 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 627 "slub_debug=-". 628 629config SLUB_STATS 630 default n 631 bool "Enable SLUB performance statistics" 632 depends on SLUB && SYSFS 633 help 634 SLUB statistics are useful to debug SLUBs allocation behavior in 635 order find ways to optimize the allocator. This should never be 636 enabled for production use since keeping statistics slows down 637 the allocator by a few percentage points. The slabinfo command 638 supports the determination of the most active slabs to figure 639 out which slabs are relevant to a particular load. 640 Try running: slabinfo -DA 641 642config HAVE_DEBUG_KMEMLEAK 643 bool 644 645config DEBUG_KMEMLEAK 646 bool "Kernel memory leak detector" 647 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 648 select DEBUG_FS 649 select STACKTRACE if STACKTRACE_SUPPORT 650 select KALLSYMS 651 select CRC32 652 help 653 Say Y here if you want to enable the memory leak 654 detector. The memory allocation/freeing is traced in a way 655 similar to the Boehm's conservative garbage collector, the 656 difference being that the orphan objects are not freed but 657 only shown in /sys/kernel/debug/kmemleak. Enabling this 658 feature will introduce an overhead to memory 659 allocations. See Documentation/dev-tools/kmemleak.rst for more 660 details. 661 662 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 663 of finding leaks due to the slab objects poisoning. 664 665 In order to access the kmemleak file, debugfs needs to be 666 mounted (usually at /sys/kernel/debug). 667 668config DEBUG_KMEMLEAK_MEM_POOL_SIZE 669 int "Kmemleak memory pool size" 670 depends on DEBUG_KMEMLEAK 671 range 200 1000000 672 default 16000 673 help 674 Kmemleak must track all the memory allocations to avoid 675 reporting false positives. Since memory may be allocated or 676 freed before kmemleak is fully initialised, use a static pool 677 of metadata objects to track such callbacks. After kmemleak is 678 fully initialised, this memory pool acts as an emergency one 679 if slab allocations fail. 680 681config DEBUG_KMEMLEAK_TEST 682 tristate "Simple test for the kernel memory leak detector" 683 depends on DEBUG_KMEMLEAK && m 684 help 685 This option enables a module that explicitly leaks memory. 686 687 If unsure, say N. 688 689config DEBUG_KMEMLEAK_DEFAULT_OFF 690 bool "Default kmemleak to off" 691 depends on DEBUG_KMEMLEAK 692 help 693 Say Y here to disable kmemleak by default. It can then be enabled 694 on the command line via kmemleak=on. 695 696config DEBUG_KMEMLEAK_AUTO_SCAN 697 bool "Enable kmemleak auto scan thread on boot up" 698 default y 699 depends on DEBUG_KMEMLEAK 700 help 701 Depending on the cpu, kmemleak scan may be cpu intensive and can 702 stall user tasks at times. This option enables/disables automatic 703 kmemleak scan at boot up. 704 705 Say N here to disable kmemleak auto scan thread to stop automatic 706 scanning. Disabling this option disables automatic reporting of 707 memory leaks. 708 709 If unsure, say Y. 710 711config DEBUG_STACK_USAGE 712 bool "Stack utilization instrumentation" 713 depends on DEBUG_KERNEL && !IA64 714 help 715 Enables the display of the minimum amount of free stack which each 716 task has ever had available in the sysrq-T and sysrq-P debug output. 717 718 This option will slow down process creation somewhat. 719 720config SCHED_STACK_END_CHECK 721 bool "Detect stack corruption on calls to schedule()" 722 depends on DEBUG_KERNEL 723 default n 724 help 725 This option checks for a stack overrun on calls to schedule(). 726 If the stack end location is found to be over written always panic as 727 the content of the corrupted region can no longer be trusted. 728 This is to ensure no erroneous behaviour occurs which could result in 729 data corruption or a sporadic crash at a later stage once the region 730 is examined. The runtime overhead introduced is minimal. 731 732config ARCH_HAS_DEBUG_VM_PGTABLE 733 bool 734 help 735 An architecture should select this when it can successfully 736 build and run DEBUG_VM_PGTABLE. 737 738config DEBUG_VM 739 bool "Debug VM" 740 depends on DEBUG_KERNEL 741 help 742 Enable this to turn on extended checks in the virtual-memory system 743 that may impact performance. 744 745 If unsure, say N. 746 747config DEBUG_VM_VMACACHE 748 bool "Debug VMA caching" 749 depends on DEBUG_VM 750 help 751 Enable this to turn on VMA caching debug information. Doing so 752 can cause significant overhead, so only enable it in non-production 753 environments. 754 755 If unsure, say N. 756 757config DEBUG_VM_RB 758 bool "Debug VM red-black trees" 759 depends on DEBUG_VM 760 help 761 Enable VM red-black tree debugging information and extra validations. 762 763 If unsure, say N. 764 765config DEBUG_VM_PGFLAGS 766 bool "Debug page-flags operations" 767 depends on DEBUG_VM 768 help 769 Enables extra validation on page flags operations. 770 771 If unsure, say N. 772 773config DEBUG_VM_PGTABLE 774 bool "Debug arch page table for semantics compliance" 775 depends on MMU 776 depends on ARCH_HAS_DEBUG_VM_PGTABLE 777 default y if DEBUG_VM 778 help 779 This option provides a debug method which can be used to test 780 architecture page table helper functions on various platforms in 781 verifying if they comply with expected generic MM semantics. This 782 will help architecture code in making sure that any changes or 783 new additions of these helpers still conform to expected 784 semantics of the generic MM. Platforms will have to opt in for 785 this through ARCH_HAS_DEBUG_VM_PGTABLE. 786 787 If unsure, say N. 788 789config ARCH_HAS_DEBUG_VIRTUAL 790 bool 791 792config DEBUG_VIRTUAL 793 bool "Debug VM translations" 794 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 795 help 796 Enable some costly sanity checks in virtual to page code. This can 797 catch mistakes with virt_to_page() and friends. 798 799 If unsure, say N. 800 801config DEBUG_NOMMU_REGIONS 802 bool "Debug the global anon/private NOMMU mapping region tree" 803 depends on DEBUG_KERNEL && !MMU 804 help 805 This option causes the global tree of anonymous and private mapping 806 regions to be regularly checked for invalid topology. 807 808config DEBUG_MEMORY_INIT 809 bool "Debug memory initialisation" if EXPERT 810 default !EXPERT 811 help 812 Enable this for additional checks during memory initialisation. 813 The sanity checks verify aspects of the VM such as the memory model 814 and other information provided by the architecture. Verbose 815 information will be printed at KERN_DEBUG loglevel depending 816 on the mminit_loglevel= command-line option. 817 818 If unsure, say Y 819 820config MEMORY_NOTIFIER_ERROR_INJECT 821 tristate "Memory hotplug notifier error injection module" 822 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 823 help 824 This option provides the ability to inject artificial errors to 825 memory hotplug notifier chain callbacks. It is controlled through 826 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 827 828 If the notifier call chain should be failed with some events 829 notified, write the error code to "actions/<notifier event>/error". 830 831 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 832 833 # cd /sys/kernel/debug/notifier-error-inject/memory 834 # echo -12 > actions/MEM_GOING_OFFLINE/error 835 # echo offline > /sys/devices/system/memory/memoryXXX/state 836 bash: echo: write error: Cannot allocate memory 837 838 To compile this code as a module, choose M here: the module will 839 be called memory-notifier-error-inject. 840 841 If unsure, say N. 842 843config DEBUG_PER_CPU_MAPS 844 bool "Debug access to per_cpu maps" 845 depends on DEBUG_KERNEL 846 depends on SMP 847 help 848 Say Y to verify that the per_cpu map being accessed has 849 been set up. This adds a fair amount of code to kernel memory 850 and decreases performance. 851 852 Say N if unsure. 853 854config DEBUG_HIGHMEM 855 bool "Highmem debugging" 856 depends on DEBUG_KERNEL && HIGHMEM 857 help 858 This option enables additional error checking for high memory 859 systems. Disable for production systems. 860 861config HAVE_DEBUG_STACKOVERFLOW 862 bool 863 864config DEBUG_STACKOVERFLOW 865 bool "Check for stack overflows" 866 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 867 help 868 Say Y here if you want to check for overflows of kernel, IRQ 869 and exception stacks (if your architecture uses them). This 870 option will show detailed messages if free stack space drops 871 below a certain limit. 872 873 These kinds of bugs usually occur when call-chains in the 874 kernel get too deep, especially when interrupts are 875 involved. 876 877 Use this in cases where you see apparently random memory 878 corruption, especially if it appears in 'struct thread_info' 879 880 If in doubt, say "N". 881 882source "lib/Kconfig.kasan" 883 884endmenu # "Memory Debugging" 885 886config DEBUG_SHIRQ 887 bool "Debug shared IRQ handlers" 888 depends on DEBUG_KERNEL 889 help 890 Enable this to generate a spurious interrupt just before a shared 891 interrupt handler is deregistered (generating one when registering 892 is currently disabled). Drivers need to handle this correctly. Some 893 don't and need to be caught. 894 895menu "Debug Oops, Lockups and Hangs" 896 897config PANIC_ON_OOPS 898 bool "Panic on Oops" 899 help 900 Say Y here to enable the kernel to panic when it oopses. This 901 has the same effect as setting oops=panic on the kernel command 902 line. 903 904 This feature is useful to ensure that the kernel does not do 905 anything erroneous after an oops which could result in data 906 corruption or other issues. 907 908 Say N if unsure. 909 910config PANIC_ON_OOPS_VALUE 911 int 912 range 0 1 913 default 0 if !PANIC_ON_OOPS 914 default 1 if PANIC_ON_OOPS 915 916config PANIC_TIMEOUT 917 int "panic timeout" 918 default 0 919 help 920 Set the timeout value (in seconds) until a reboot occurs when 921 the kernel panics. If n = 0, then we wait forever. A timeout 922 value n > 0 will wait n seconds before rebooting, while a timeout 923 value n < 0 will reboot immediately. 924 925config LOCKUP_DETECTOR 926 bool 927 928config SOFTLOCKUP_DETECTOR 929 bool "Detect Soft Lockups" 930 depends on DEBUG_KERNEL && !S390 931 select LOCKUP_DETECTOR 932 help 933 Say Y here to enable the kernel to act as a watchdog to detect 934 soft lockups. 935 936 Softlockups are bugs that cause the kernel to loop in kernel 937 mode for more than 20 seconds, without giving other tasks a 938 chance to run. The current stack trace is displayed upon 939 detection and the system will stay locked up. 940 941config BOOTPARAM_SOFTLOCKUP_PANIC 942 bool "Panic (Reboot) On Soft Lockups" 943 depends on SOFTLOCKUP_DETECTOR 944 help 945 Say Y here to enable the kernel to panic on "soft lockups", 946 which are bugs that cause the kernel to loop in kernel 947 mode for more than 20 seconds (configurable using the watchdog_thresh 948 sysctl), without giving other tasks a chance to run. 949 950 The panic can be used in combination with panic_timeout, 951 to cause the system to reboot automatically after a 952 lockup has been detected. This feature is useful for 953 high-availability systems that have uptime guarantees and 954 where a lockup must be resolved ASAP. 955 956 Say N if unsure. 957 958config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 959 int 960 depends on SOFTLOCKUP_DETECTOR 961 range 0 1 962 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 963 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 964 965config HARDLOCKUP_DETECTOR_PERF 966 bool 967 select SOFTLOCKUP_DETECTOR 968 969# 970# Enables a timestamp based low pass filter to compensate for perf based 971# hard lockup detection which runs too fast due to turbo modes. 972# 973config HARDLOCKUP_CHECK_TIMESTAMP 974 bool 975 976# 977# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard 978# lockup detector rather than the perf based detector. 979# 980config HARDLOCKUP_DETECTOR 981 bool "Detect Hard Lockups" 982 depends on DEBUG_KERNEL && !S390 983 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH 984 select LOCKUP_DETECTOR 985 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF 986 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH 987 help 988 Say Y here to enable the kernel to act as a watchdog to detect 989 hard lockups. 990 991 Hardlockups are bugs that cause the CPU to loop in kernel mode 992 for more than 10 seconds, without letting other interrupts have a 993 chance to run. The current stack trace is displayed upon detection 994 and the system will stay locked up. 995 996config BOOTPARAM_HARDLOCKUP_PANIC 997 bool "Panic (Reboot) On Hard Lockups" 998 depends on HARDLOCKUP_DETECTOR 999 help 1000 Say Y here to enable the kernel to panic on "hard lockups", 1001 which are bugs that cause the kernel to loop in kernel 1002 mode with interrupts disabled for more than 10 seconds (configurable 1003 using the watchdog_thresh sysctl). 1004 1005 Say N if unsure. 1006 1007config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 1008 int 1009 depends on HARDLOCKUP_DETECTOR 1010 range 0 1 1011 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 1012 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 1013 1014config DETECT_HUNG_TASK 1015 bool "Detect Hung Tasks" 1016 depends on DEBUG_KERNEL 1017 default SOFTLOCKUP_DETECTOR 1018 help 1019 Say Y here to enable the kernel to detect "hung tasks", 1020 which are bugs that cause the task to be stuck in 1021 uninterruptible "D" state indefinitely. 1022 1023 When a hung task is detected, the kernel will print the 1024 current stack trace (which you should report), but the 1025 task will stay in uninterruptible state. If lockdep is 1026 enabled then all held locks will also be reported. This 1027 feature has negligible overhead. 1028 1029config DEFAULT_HUNG_TASK_TIMEOUT 1030 int "Default timeout for hung task detection (in seconds)" 1031 depends on DETECT_HUNG_TASK 1032 default 120 1033 help 1034 This option controls the default timeout (in seconds) used 1035 to determine when a task has become non-responsive and should 1036 be considered hung. 1037 1038 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 1039 sysctl or by writing a value to 1040 /proc/sys/kernel/hung_task_timeout_secs. 1041 1042 A timeout of 0 disables the check. The default is two minutes. 1043 Keeping the default should be fine in most cases. 1044 1045config BOOTPARAM_HUNG_TASK_PANIC 1046 bool "Panic (Reboot) On Hung Tasks" 1047 depends on DETECT_HUNG_TASK 1048 help 1049 Say Y here to enable the kernel to panic on "hung tasks", 1050 which are bugs that cause the kernel to leave a task stuck 1051 in uninterruptible "D" state. 1052 1053 The panic can be used in combination with panic_timeout, 1054 to cause the system to reboot automatically after a 1055 hung task has been detected. This feature is useful for 1056 high-availability systems that have uptime guarantees and 1057 where a hung tasks must be resolved ASAP. 1058 1059 Say N if unsure. 1060 1061config BOOTPARAM_HUNG_TASK_PANIC_VALUE 1062 int 1063 depends on DETECT_HUNG_TASK 1064 range 0 1 1065 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 1066 default 1 if BOOTPARAM_HUNG_TASK_PANIC 1067 1068config WQ_WATCHDOG 1069 bool "Detect Workqueue Stalls" 1070 depends on DEBUG_KERNEL 1071 help 1072 Say Y here to enable stall detection on workqueues. If a 1073 worker pool doesn't make forward progress on a pending work 1074 item for over a given amount of time, 30s by default, a 1075 warning message is printed along with dump of workqueue 1076 state. This can be configured through kernel parameter 1077 "workqueue.watchdog_thresh" and its sysfs counterpart. 1078 1079config TEST_LOCKUP 1080 tristate "Test module to generate lockups" 1081 depends on m 1082 help 1083 This builds the "test_lockup" module that helps to make sure 1084 that watchdogs and lockup detectors are working properly. 1085 1086 Depending on module parameters it could emulate soft or hard 1087 lockup, "hung task", or locking arbitrary lock for a long time. 1088 Also it could generate series of lockups with cooling-down periods. 1089 1090 If unsure, say N. 1091 1092endmenu # "Debug lockups and hangs" 1093 1094menu "Scheduler Debugging" 1095 1096config SCHED_DEBUG 1097 bool "Collect scheduler debugging info" 1098 depends on DEBUG_KERNEL && PROC_FS 1099 default y 1100 help 1101 If you say Y here, the /proc/sched_debug file will be provided 1102 that can help debug the scheduler. The runtime overhead of this 1103 option is minimal. 1104 1105config SCHED_INFO 1106 bool 1107 default n 1108 1109config SCHEDSTATS 1110 bool "Collect scheduler statistics" 1111 depends on DEBUG_KERNEL && PROC_FS 1112 select SCHED_INFO 1113 help 1114 If you say Y here, additional code will be inserted into the 1115 scheduler and related routines to collect statistics about 1116 scheduler behavior and provide them in /proc/schedstat. These 1117 stats may be useful for both tuning and debugging the scheduler 1118 If you aren't debugging the scheduler or trying to tune a specific 1119 application, you can say N to avoid the very slight overhead 1120 this adds. 1121 1122endmenu 1123 1124config DEBUG_TIMEKEEPING 1125 bool "Enable extra timekeeping sanity checking" 1126 help 1127 This option will enable additional timekeeping sanity checks 1128 which may be helpful when diagnosing issues where timekeeping 1129 problems are suspected. 1130 1131 This may include checks in the timekeeping hotpaths, so this 1132 option may have a (very small) performance impact to some 1133 workloads. 1134 1135 If unsure, say N. 1136 1137config DEBUG_PREEMPT 1138 bool "Debug preemptible kernel" 1139 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT 1140 default y 1141 help 1142 If you say Y here then the kernel will use a debug variant of the 1143 commonly used smp_processor_id() function and will print warnings 1144 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1145 will detect preemption count underflows. 1146 1147menu "Lock Debugging (spinlocks, mutexes, etc...)" 1148 1149config LOCK_DEBUGGING_SUPPORT 1150 bool 1151 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1152 default y 1153 1154config PROVE_LOCKING 1155 bool "Lock debugging: prove locking correctness" 1156 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1157 select LOCKDEP 1158 select DEBUG_SPINLOCK 1159 select DEBUG_MUTEXES 1160 select DEBUG_RT_MUTEXES if RT_MUTEXES 1161 select DEBUG_RWSEMS 1162 select DEBUG_WW_MUTEX_SLOWPATH 1163 select DEBUG_LOCK_ALLOC 1164 select PREEMPT_COUNT if !ARCH_NO_PREEMPT 1165 select TRACE_IRQFLAGS 1166 default n 1167 help 1168 This feature enables the kernel to prove that all locking 1169 that occurs in the kernel runtime is mathematically 1170 correct: that under no circumstance could an arbitrary (and 1171 not yet triggered) combination of observed locking 1172 sequences (on an arbitrary number of CPUs, running an 1173 arbitrary number of tasks and interrupt contexts) cause a 1174 deadlock. 1175 1176 In short, this feature enables the kernel to report locking 1177 related deadlocks before they actually occur. 1178 1179 The proof does not depend on how hard and complex a 1180 deadlock scenario would be to trigger: how many 1181 participant CPUs, tasks and irq-contexts would be needed 1182 for it to trigger. The proof also does not depend on 1183 timing: if a race and a resulting deadlock is possible 1184 theoretically (no matter how unlikely the race scenario 1185 is), it will be proven so and will immediately be 1186 reported by the kernel (once the event is observed that 1187 makes the deadlock theoretically possible). 1188 1189 If a deadlock is impossible (i.e. the locking rules, as 1190 observed by the kernel, are mathematically correct), the 1191 kernel reports nothing. 1192 1193 NOTE: this feature can also be enabled for rwlocks, mutexes 1194 and rwsems - in which case all dependencies between these 1195 different locking variants are observed and mapped too, and 1196 the proof of observed correctness is also maintained for an 1197 arbitrary combination of these separate locking variants. 1198 1199 For more details, see Documentation/locking/lockdep-design.rst. 1200 1201config PROVE_RAW_LOCK_NESTING 1202 bool "Enable raw_spinlock - spinlock nesting checks" 1203 depends on PROVE_LOCKING 1204 default n 1205 help 1206 Enable the raw_spinlock vs. spinlock nesting checks which ensure 1207 that the lock nesting rules for PREEMPT_RT enabled kernels are 1208 not violated. 1209 1210 NOTE: There are known nesting problems. So if you enable this 1211 option expect lockdep splats until these problems have been fully 1212 addressed which is work in progress. This config switch allows to 1213 identify and analyze these problems. It will be removed and the 1214 check permanentely enabled once the main issues have been fixed. 1215 1216 If unsure, select N. 1217 1218config LOCK_STAT 1219 bool "Lock usage statistics" 1220 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1221 select LOCKDEP 1222 select DEBUG_SPINLOCK 1223 select DEBUG_MUTEXES 1224 select DEBUG_RT_MUTEXES if RT_MUTEXES 1225 select DEBUG_LOCK_ALLOC 1226 default n 1227 help 1228 This feature enables tracking lock contention points 1229 1230 For more details, see Documentation/locking/lockstat.rst 1231 1232 This also enables lock events required by "perf lock", 1233 subcommand of perf. 1234 If you want to use "perf lock", you also need to turn on 1235 CONFIG_EVENT_TRACING. 1236 1237 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1238 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1239 1240config DEBUG_RT_MUTEXES 1241 bool "RT Mutex debugging, deadlock detection" 1242 depends on DEBUG_KERNEL && RT_MUTEXES 1243 help 1244 This allows rt mutex semantics violations and rt mutex related 1245 deadlocks (lockups) to be detected and reported automatically. 1246 1247config DEBUG_SPINLOCK 1248 bool "Spinlock and rw-lock debugging: basic checks" 1249 depends on DEBUG_KERNEL 1250 select UNINLINE_SPIN_UNLOCK 1251 help 1252 Say Y here and build SMP to catch missing spinlock initialization 1253 and certain other kinds of spinlock errors commonly made. This is 1254 best used in conjunction with the NMI watchdog so that spinlock 1255 deadlocks are also debuggable. 1256 1257config DEBUG_MUTEXES 1258 bool "Mutex debugging: basic checks" 1259 depends on DEBUG_KERNEL 1260 help 1261 This feature allows mutex semantics violations to be detected and 1262 reported. 1263 1264config DEBUG_WW_MUTEX_SLOWPATH 1265 bool "Wait/wound mutex debugging: Slowpath testing" 1266 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1267 select DEBUG_LOCK_ALLOC 1268 select DEBUG_SPINLOCK 1269 select DEBUG_MUTEXES 1270 help 1271 This feature enables slowpath testing for w/w mutex users by 1272 injecting additional -EDEADLK wound/backoff cases. Together with 1273 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1274 will test all possible w/w mutex interface abuse with the 1275 exception of simply not acquiring all the required locks. 1276 Note that this feature can introduce significant overhead, so 1277 it really should not be enabled in a production or distro kernel, 1278 even a debug kernel. If you are a driver writer, enable it. If 1279 you are a distro, do not. 1280 1281config DEBUG_RWSEMS 1282 bool "RW Semaphore debugging: basic checks" 1283 depends on DEBUG_KERNEL 1284 help 1285 This debugging feature allows mismatched rw semaphore locks 1286 and unlocks to be detected and reported. 1287 1288config DEBUG_LOCK_ALLOC 1289 bool "Lock debugging: detect incorrect freeing of live locks" 1290 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1291 select DEBUG_SPINLOCK 1292 select DEBUG_MUTEXES 1293 select DEBUG_RT_MUTEXES if RT_MUTEXES 1294 select LOCKDEP 1295 help 1296 This feature will check whether any held lock (spinlock, rwlock, 1297 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1298 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1299 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1300 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1301 held during task exit. 1302 1303config LOCKDEP 1304 bool 1305 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1306 select STACKTRACE 1307 select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86 1308 select KALLSYMS 1309 select KALLSYMS_ALL 1310 1311config LOCKDEP_SMALL 1312 bool 1313 1314config DEBUG_LOCKDEP 1315 bool "Lock dependency engine debugging" 1316 depends on DEBUG_KERNEL && LOCKDEP 1317 help 1318 If you say Y here, the lock dependency engine will do 1319 additional runtime checks to debug itself, at the price 1320 of more runtime overhead. 1321 1322config DEBUG_ATOMIC_SLEEP 1323 bool "Sleep inside atomic section checking" 1324 select PREEMPT_COUNT 1325 depends on DEBUG_KERNEL 1326 depends on !ARCH_NO_PREEMPT 1327 help 1328 If you say Y here, various routines which may sleep will become very 1329 noisy if they are called inside atomic sections: when a spinlock is 1330 held, inside an rcu read side critical section, inside preempt disabled 1331 sections, inside an interrupt, etc... 1332 1333config DEBUG_LOCKING_API_SELFTESTS 1334 bool "Locking API boot-time self-tests" 1335 depends on DEBUG_KERNEL 1336 help 1337 Say Y here if you want the kernel to run a short self-test during 1338 bootup. The self-test checks whether common types of locking bugs 1339 are detected by debugging mechanisms or not. (if you disable 1340 lock debugging then those bugs wont be detected of course.) 1341 The following locking APIs are covered: spinlocks, rwlocks, 1342 mutexes and rwsems. 1343 1344config LOCK_TORTURE_TEST 1345 tristate "torture tests for locking" 1346 depends on DEBUG_KERNEL 1347 select TORTURE_TEST 1348 help 1349 This option provides a kernel module that runs torture tests 1350 on kernel locking primitives. The kernel module may be built 1351 after the fact on the running kernel to be tested, if desired. 1352 1353 Say Y here if you want kernel locking-primitive torture tests 1354 to be built into the kernel. 1355 Say M if you want these torture tests to build as a module. 1356 Say N if you are unsure. 1357 1358config WW_MUTEX_SELFTEST 1359 tristate "Wait/wound mutex selftests" 1360 help 1361 This option provides a kernel module that runs tests on the 1362 on the struct ww_mutex locking API. 1363 1364 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1365 with this test harness. 1366 1367 Say M if you want these self tests to build as a module. 1368 Say N if you are unsure. 1369 1370endmenu # lock debugging 1371 1372config TRACE_IRQFLAGS 1373 depends on TRACE_IRQFLAGS_SUPPORT 1374 bool 1375 help 1376 Enables hooks to interrupt enabling and disabling for 1377 either tracing or lock debugging. 1378 1379config TRACE_IRQFLAGS_NMI 1380 def_bool y 1381 depends on TRACE_IRQFLAGS 1382 depends on TRACE_IRQFLAGS_NMI_SUPPORT 1383 1384config STACKTRACE 1385 bool "Stack backtrace support" 1386 depends on STACKTRACE_SUPPORT 1387 help 1388 This option causes the kernel to create a /proc/pid/stack for 1389 every process, showing its current stack trace. 1390 It is also used by various kernel debugging features that require 1391 stack trace generation. 1392 1393config WARN_ALL_UNSEEDED_RANDOM 1394 bool "Warn for all uses of unseeded randomness" 1395 default n 1396 help 1397 Some parts of the kernel contain bugs relating to their use of 1398 cryptographically secure random numbers before it's actually possible 1399 to generate those numbers securely. This setting ensures that these 1400 flaws don't go unnoticed, by enabling a message, should this ever 1401 occur. This will allow people with obscure setups to know when things 1402 are going wrong, so that they might contact developers about fixing 1403 it. 1404 1405 Unfortunately, on some models of some architectures getting 1406 a fully seeded CRNG is extremely difficult, and so this can 1407 result in dmesg getting spammed for a surprisingly long 1408 time. This is really bad from a security perspective, and 1409 so architecture maintainers really need to do what they can 1410 to get the CRNG seeded sooner after the system is booted. 1411 However, since users cannot do anything actionable to 1412 address this, by default the kernel will issue only a single 1413 warning for the first use of unseeded randomness. 1414 1415 Say Y here if you want to receive warnings for all uses of 1416 unseeded randomness. This will be of use primarily for 1417 those developers interested in improving the security of 1418 Linux kernels running on their architecture (or 1419 subarchitecture). 1420 1421config DEBUG_KOBJECT 1422 bool "kobject debugging" 1423 depends on DEBUG_KERNEL 1424 help 1425 If you say Y here, some extra kobject debugging messages will be sent 1426 to the syslog. 1427 1428config DEBUG_KOBJECT_RELEASE 1429 bool "kobject release debugging" 1430 depends on DEBUG_OBJECTS_TIMERS 1431 help 1432 kobjects are reference counted objects. This means that their 1433 last reference count put is not predictable, and the kobject can 1434 live on past the point at which a driver decides to drop it's 1435 initial reference to the kobject gained on allocation. An 1436 example of this would be a struct device which has just been 1437 unregistered. 1438 1439 However, some buggy drivers assume that after such an operation, 1440 the memory backing the kobject can be immediately freed. This 1441 goes completely against the principles of a refcounted object. 1442 1443 If you say Y here, the kernel will delay the release of kobjects 1444 on the last reference count to improve the visibility of this 1445 kind of kobject release bug. 1446 1447config HAVE_DEBUG_BUGVERBOSE 1448 bool 1449 1450menu "Debug kernel data structures" 1451 1452config DEBUG_LIST 1453 bool "Debug linked list manipulation" 1454 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION 1455 help 1456 Enable this to turn on extended checks in the linked-list 1457 walking routines. 1458 1459 If unsure, say N. 1460 1461config DEBUG_PLIST 1462 bool "Debug priority linked list manipulation" 1463 depends on DEBUG_KERNEL 1464 help 1465 Enable this to turn on extended checks in the priority-ordered 1466 linked-list (plist) walking routines. This checks the entire 1467 list multiple times during each manipulation. 1468 1469 If unsure, say N. 1470 1471config DEBUG_SG 1472 bool "Debug SG table operations" 1473 depends on DEBUG_KERNEL 1474 help 1475 Enable this to turn on checks on scatter-gather tables. This can 1476 help find problems with drivers that do not properly initialize 1477 their sg tables. 1478 1479 If unsure, say N. 1480 1481config DEBUG_NOTIFIERS 1482 bool "Debug notifier call chains" 1483 depends on DEBUG_KERNEL 1484 help 1485 Enable this to turn on sanity checking for notifier call chains. 1486 This is most useful for kernel developers to make sure that 1487 modules properly unregister themselves from notifier chains. 1488 This is a relatively cheap check but if you care about maximum 1489 performance, say N. 1490 1491config BUG_ON_DATA_CORRUPTION 1492 bool "Trigger a BUG when data corruption is detected" 1493 select DEBUG_LIST 1494 help 1495 Select this option if the kernel should BUG when it encounters 1496 data corruption in kernel memory structures when they get checked 1497 for validity. 1498 1499 If unsure, say N. 1500 1501endmenu 1502 1503config DEBUG_CREDENTIALS 1504 bool "Debug credential management" 1505 depends on DEBUG_KERNEL 1506 help 1507 Enable this to turn on some debug checking for credential 1508 management. The additional code keeps track of the number of 1509 pointers from task_structs to any given cred struct, and checks to 1510 see that this number never exceeds the usage count of the cred 1511 struct. 1512 1513 Furthermore, if SELinux is enabled, this also checks that the 1514 security pointer in the cred struct is never seen to be invalid. 1515 1516 If unsure, say N. 1517 1518source "kernel/rcu/Kconfig.debug" 1519 1520config DEBUG_WQ_FORCE_RR_CPU 1521 bool "Force round-robin CPU selection for unbound work items" 1522 depends on DEBUG_KERNEL 1523 default n 1524 help 1525 Workqueue used to implicitly guarantee that work items queued 1526 without explicit CPU specified are put on the local CPU. This 1527 guarantee is no longer true and while local CPU is still 1528 preferred work items may be put on foreign CPUs. Kernel 1529 parameter "workqueue.debug_force_rr_cpu" is added to force 1530 round-robin CPU selection to flush out usages which depend on the 1531 now broken guarantee. This config option enables the debug 1532 feature by default. When enabled, memory and cache locality will 1533 be impacted. 1534 1535config DEBUG_BLOCK_EXT_DEVT 1536 bool "Force extended block device numbers and spread them" 1537 depends on DEBUG_KERNEL 1538 depends on BLOCK 1539 default n 1540 help 1541 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1542 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1543 YOU ARE DOING. Distros, please enable this and fix whatever 1544 is broken. 1545 1546 Conventionally, block device numbers are allocated from 1547 predetermined contiguous area. However, extended block area 1548 may introduce non-contiguous block device numbers. This 1549 option forces most block device numbers to be allocated from 1550 the extended space and spreads them to discover kernel or 1551 userland code paths which assume predetermined contiguous 1552 device number allocation. 1553 1554 Note that turning on this debug option shuffles all the 1555 device numbers for all IDE and SCSI devices including libata 1556 ones, so root partition specified using device number 1557 directly (via rdev or root=MAJ:MIN) won't work anymore. 1558 Textual device names (root=/dev/sdXn) will continue to work. 1559 1560 Say N if you are unsure. 1561 1562config CPU_HOTPLUG_STATE_CONTROL 1563 bool "Enable CPU hotplug state control" 1564 depends on DEBUG_KERNEL 1565 depends on HOTPLUG_CPU 1566 default n 1567 help 1568 Allows to write steps between "offline" and "online" to the CPUs 1569 sysfs target file so states can be stepped granular. This is a debug 1570 option for now as the hotplug machinery cannot be stopped and 1571 restarted at arbitrary points yet. 1572 1573 Say N if your are unsure. 1574 1575config LATENCYTOP 1576 bool "Latency measuring infrastructure" 1577 depends on DEBUG_KERNEL 1578 depends on STACKTRACE_SUPPORT 1579 depends on PROC_FS 1580 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1581 select KALLSYMS 1582 select KALLSYMS_ALL 1583 select STACKTRACE 1584 select SCHEDSTATS 1585 select SCHED_DEBUG 1586 help 1587 Enable this option if you want to use the LatencyTOP tool 1588 to find out which userspace is blocking on what kernel operations. 1589 1590source "kernel/trace/Kconfig" 1591 1592config PROVIDE_OHCI1394_DMA_INIT 1593 bool "Remote debugging over FireWire early on boot" 1594 depends on PCI && X86 1595 help 1596 If you want to debug problems which hang or crash the kernel early 1597 on boot and the crashing machine has a FireWire port, you can use 1598 this feature to remotely access the memory of the crashed machine 1599 over FireWire. This employs remote DMA as part of the OHCI1394 1600 specification which is now the standard for FireWire controllers. 1601 1602 With remote DMA, you can monitor the printk buffer remotely using 1603 firescope and access all memory below 4GB using fireproxy from gdb. 1604 Even controlling a kernel debugger is possible using remote DMA. 1605 1606 Usage: 1607 1608 If ohci1394_dma=early is used as boot parameter, it will initialize 1609 all OHCI1394 controllers which are found in the PCI config space. 1610 1611 As all changes to the FireWire bus such as enabling and disabling 1612 devices cause a bus reset and thereby disable remote DMA for all 1613 devices, be sure to have the cable plugged and FireWire enabled on 1614 the debugging host before booting the debug target for debugging. 1615 1616 This code (~1k) is freed after boot. By then, the firewire stack 1617 in charge of the OHCI-1394 controllers should be used instead. 1618 1619 See Documentation/core-api/debugging-via-ohci1394.rst for more information. 1620 1621source "samples/Kconfig" 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 FAULT_INJECTION_USERCOPY 1772 bool "Fault injection capability for usercopy functions" 1773 depends on FAULT_INJECTION 1774 help 1775 Provides fault-injection capability to inject failures 1776 in usercopy functions (copy_from_user(), get_user(), ...). 1777 1778config FAIL_MAKE_REQUEST 1779 bool "Fault-injection capability for disk IO" 1780 depends on FAULT_INJECTION && BLOCK 1781 help 1782 Provide fault-injection capability for disk IO. 1783 1784config FAIL_IO_TIMEOUT 1785 bool "Fault-injection capability for faking disk interrupts" 1786 depends on FAULT_INJECTION && BLOCK 1787 help 1788 Provide fault-injection capability on end IO handling. This 1789 will make the block layer "forget" an interrupt as configured, 1790 thus exercising the error handling. 1791 1792 Only works with drivers that use the generic timeout handling, 1793 for others it wont do anything. 1794 1795config FAIL_FUTEX 1796 bool "Fault-injection capability for futexes" 1797 select DEBUG_FS 1798 depends on FAULT_INJECTION && FUTEX 1799 help 1800 Provide fault-injection capability for futexes. 1801 1802config FAULT_INJECTION_DEBUG_FS 1803 bool "Debugfs entries for fault-injection capabilities" 1804 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1805 help 1806 Enable configuration of fault-injection capabilities via debugfs. 1807 1808config FAIL_FUNCTION 1809 bool "Fault-injection capability for functions" 1810 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION 1811 help 1812 Provide function-based fault-injection capability. 1813 This will allow you to override a specific function with a return 1814 with given return value. As a result, function caller will see 1815 an error value and have to handle it. This is useful to test the 1816 error handling in various subsystems. 1817 1818config FAIL_MMC_REQUEST 1819 bool "Fault-injection capability for MMC IO" 1820 depends on FAULT_INJECTION_DEBUG_FS && MMC 1821 help 1822 Provide fault-injection capability for MMC IO. 1823 This will make the mmc core return data errors. This is 1824 useful to test the error handling in the mmc block device 1825 and to test how the mmc host driver handles retries from 1826 the block device. 1827 1828config FAULT_INJECTION_STACKTRACE_FILTER 1829 bool "stacktrace filter for fault-injection capabilities" 1830 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1831 depends on !X86_64 1832 select STACKTRACE 1833 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1834 help 1835 Provide stacktrace filter for fault-injection capabilities 1836 1837config ARCH_HAS_KCOV 1838 bool 1839 help 1840 An architecture should select this when it can successfully 1841 build and run with CONFIG_KCOV. This typically requires 1842 disabling instrumentation for some early boot code. 1843 1844config CC_HAS_SANCOV_TRACE_PC 1845 def_bool $(cc-option,-fsanitize-coverage=trace-pc) 1846 1847 1848config KCOV 1849 bool "Code coverage for fuzzing" 1850 depends on ARCH_HAS_KCOV 1851 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS 1852 select DEBUG_FS 1853 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC 1854 help 1855 KCOV exposes kernel code coverage information in a form suitable 1856 for coverage-guided fuzzing (randomized testing). 1857 1858 If RANDOMIZE_BASE is enabled, PC values will not be stable across 1859 different machines and across reboots. If you need stable PC values, 1860 disable RANDOMIZE_BASE. 1861 1862 For more details, see Documentation/dev-tools/kcov.rst. 1863 1864config KCOV_ENABLE_COMPARISONS 1865 bool "Enable comparison operands collection by KCOV" 1866 depends on KCOV 1867 depends on $(cc-option,-fsanitize-coverage=trace-cmp) 1868 help 1869 KCOV also exposes operands of every comparison in the instrumented 1870 code along with operand sizes and PCs of the comparison instructions. 1871 These operands can be used by fuzzing engines to improve the quality 1872 of fuzzing coverage. 1873 1874config KCOV_INSTRUMENT_ALL 1875 bool "Instrument all code by default" 1876 depends on KCOV 1877 default y 1878 help 1879 If you are doing generic system call fuzzing (like e.g. syzkaller), 1880 then you will want to instrument the whole kernel and you should 1881 say y here. If you are doing more targeted fuzzing (like e.g. 1882 filesystem fuzzing with AFL) then you will want to enable coverage 1883 for more specific subsets of files, and should say n here. 1884 1885config KCOV_IRQ_AREA_SIZE 1886 hex "Size of interrupt coverage collection area in words" 1887 depends on KCOV 1888 default 0x40000 1889 help 1890 KCOV uses preallocated per-cpu areas to collect coverage from 1891 soft interrupts. This specifies the size of those areas in the 1892 number of unsigned long words. 1893 1894menuconfig RUNTIME_TESTING_MENU 1895 bool "Runtime Testing" 1896 def_bool y 1897 1898if RUNTIME_TESTING_MENU 1899 1900config LKDTM 1901 tristate "Linux Kernel Dump Test Tool Module" 1902 depends on DEBUG_FS 1903 help 1904 This module enables testing of the different dumping mechanisms by 1905 inducing system failures at predefined crash points. 1906 If you don't need it: say N 1907 Choose M here to compile this code as a module. The module will be 1908 called lkdtm. 1909 1910 Documentation on how to use the module can be found in 1911 Documentation/fault-injection/provoke-crashes.rst 1912 1913config TEST_LIST_SORT 1914 tristate "Linked list sorting test" 1915 depends on DEBUG_KERNEL || m 1916 help 1917 Enable this to turn on 'list_sort()' function test. This test is 1918 executed only once during system boot (so affects only boot time), 1919 or at module load time. 1920 1921 If unsure, say N. 1922 1923config TEST_MIN_HEAP 1924 tristate "Min heap test" 1925 depends on DEBUG_KERNEL || m 1926 help 1927 Enable this to turn on min heap function tests. This test is 1928 executed only once during system boot (so affects only boot time), 1929 or at module load time. 1930 1931 If unsure, say N. 1932 1933config TEST_SORT 1934 tristate "Array-based sort test" 1935 depends on DEBUG_KERNEL || m 1936 help 1937 This option enables the self-test function of 'sort()' at boot, 1938 or at module load time. 1939 1940 If unsure, say N. 1941 1942config KPROBES_SANITY_TEST 1943 bool "Kprobes sanity tests" 1944 depends on DEBUG_KERNEL 1945 depends on KPROBES 1946 help 1947 This option provides for testing basic kprobes functionality on 1948 boot. Samples of kprobe and kretprobe are inserted and 1949 verified for functionality. 1950 1951 Say N if you are unsure. 1952 1953config BACKTRACE_SELF_TEST 1954 tristate "Self test for the backtrace code" 1955 depends on DEBUG_KERNEL 1956 help 1957 This option provides a kernel module that can be used to test 1958 the kernel stack backtrace code. This option is not useful 1959 for distributions or general kernels, but only for kernel 1960 developers working on architecture code. 1961 1962 Note that if you want to also test saved backtraces, you will 1963 have to enable STACKTRACE as well. 1964 1965 Say N if you are unsure. 1966 1967config RBTREE_TEST 1968 tristate "Red-Black tree test" 1969 depends on DEBUG_KERNEL 1970 help 1971 A benchmark measuring the performance of the rbtree library. 1972 Also includes rbtree invariant checks. 1973 1974config REED_SOLOMON_TEST 1975 tristate "Reed-Solomon library test" 1976 depends on DEBUG_KERNEL || m 1977 select REED_SOLOMON 1978 select REED_SOLOMON_ENC16 1979 select REED_SOLOMON_DEC16 1980 help 1981 This option enables the self-test function of rslib at boot, 1982 or at module load time. 1983 1984 If unsure, say N. 1985 1986config INTERVAL_TREE_TEST 1987 tristate "Interval tree test" 1988 depends on DEBUG_KERNEL 1989 select INTERVAL_TREE 1990 help 1991 A benchmark measuring the performance of the interval tree library 1992 1993config PERCPU_TEST 1994 tristate "Per cpu operations test" 1995 depends on m && DEBUG_KERNEL 1996 help 1997 Enable this option to build test module which validates per-cpu 1998 operations. 1999 2000 If unsure, say N. 2001 2002config ATOMIC64_SELFTEST 2003 tristate "Perform an atomic64_t self-test" 2004 help 2005 Enable this option to test the atomic64_t functions at boot or 2006 at module load time. 2007 2008 If unsure, say N. 2009 2010config ASYNC_RAID6_TEST 2011 tristate "Self test for hardware accelerated raid6 recovery" 2012 depends on ASYNC_RAID6_RECOV 2013 select ASYNC_MEMCPY 2014 help 2015 This is a one-shot self test that permutes through the 2016 recovery of all the possible two disk failure scenarios for a 2017 N-disk array. Recovery is performed with the asynchronous 2018 raid6 recovery routines, and will optionally use an offload 2019 engine if one is available. 2020 2021 If unsure, say N. 2022 2023config TEST_HEXDUMP 2024 tristate "Test functions located in the hexdump module at runtime" 2025 2026config TEST_STRING_HELPERS 2027 tristate "Test functions located in the string_helpers module at runtime" 2028 2029config TEST_STRSCPY 2030 tristate "Test strscpy*() family of functions at runtime" 2031 2032config TEST_KSTRTOX 2033 tristate "Test kstrto*() family of functions at runtime" 2034 2035config TEST_PRINTF 2036 tristate "Test printf() family of functions at runtime" 2037 2038config TEST_BITMAP 2039 tristate "Test bitmap_*() family of functions at runtime" 2040 help 2041 Enable this option to test the bitmap functions at boot. 2042 2043 If unsure, say N. 2044 2045config TEST_BITFIELD 2046 tristate "Test bitfield functions at runtime" 2047 help 2048 Enable this option to test the bitfield functions at boot. 2049 2050 If unsure, say N. 2051 2052config TEST_UUID 2053 tristate "Test functions located in the uuid module at runtime" 2054 2055config TEST_XARRAY 2056 tristate "Test the XArray code at runtime" 2057 2058config TEST_OVERFLOW 2059 tristate "Test check_*_overflow() functions at runtime" 2060 2061config TEST_RHASHTABLE 2062 tristate "Perform selftest on resizable hash table" 2063 help 2064 Enable this option to test the rhashtable functions at boot. 2065 2066 If unsure, say N. 2067 2068config TEST_HASH 2069 tristate "Perform selftest on hash functions" 2070 help 2071 Enable this option to test the kernel's integer (<linux/hash.h>), 2072 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>) 2073 hash functions on boot (or module load). 2074 2075 This is intended to help people writing architecture-specific 2076 optimized versions. If unsure, say N. 2077 2078config TEST_IDA 2079 tristate "Perform selftest on IDA functions" 2080 2081config TEST_PARMAN 2082 tristate "Perform selftest on priority array manager" 2083 depends on PARMAN 2084 help 2085 Enable this option to test priority array manager on boot 2086 (or module load). 2087 2088 If unsure, say N. 2089 2090config TEST_IRQ_TIMINGS 2091 bool "IRQ timings selftest" 2092 depends on IRQ_TIMINGS 2093 help 2094 Enable this option to test the irq timings code on boot. 2095 2096 If unsure, say N. 2097 2098config TEST_LKM 2099 tristate "Test module loading with 'hello world' module" 2100 depends on m 2101 help 2102 This builds the "test_module" module that emits "Hello, world" 2103 on printk when loaded. It is designed to be used for basic 2104 evaluation of the module loading subsystem (for example when 2105 validating module verification). It lacks any extra dependencies, 2106 and will not normally be loaded by the system unless explicitly 2107 requested by name. 2108 2109 If unsure, say N. 2110 2111config TEST_BITOPS 2112 tristate "Test module for compilation of bitops operations" 2113 depends on m 2114 help 2115 This builds the "test_bitops" module that is much like the 2116 TEST_LKM module except that it does a basic exercise of the 2117 set/clear_bit macros and get_count_order/long to make sure there are 2118 no compiler warnings from C=1 sparse checker or -Wextra 2119 compilations. It has no dependencies and doesn't run or load unless 2120 explicitly requested by name. for example: modprobe test_bitops. 2121 2122 If unsure, say N. 2123 2124config TEST_VMALLOC 2125 tristate "Test module for stress/performance analysis of vmalloc allocator" 2126 default n 2127 depends on MMU 2128 depends on m 2129 help 2130 This builds the "test_vmalloc" module that should be used for 2131 stress and performance analysis. So, any new change for vmalloc 2132 subsystem can be evaluated from performance and stability point 2133 of view. 2134 2135 If unsure, say N. 2136 2137config TEST_USER_COPY 2138 tristate "Test user/kernel boundary protections" 2139 depends on m 2140 help 2141 This builds the "test_user_copy" module that runs sanity checks 2142 on the copy_to/from_user infrastructure, making sure basic 2143 user/kernel boundary testing is working. If it fails to load, 2144 a regression has been detected in the user/kernel memory boundary 2145 protections. 2146 2147 If unsure, say N. 2148 2149config TEST_BPF 2150 tristate "Test BPF filter functionality" 2151 depends on m && NET 2152 help 2153 This builds the "test_bpf" module that runs various test vectors 2154 against the BPF interpreter or BPF JIT compiler depending on the 2155 current setting. This is in particular useful for BPF JIT compiler 2156 development, but also to run regression tests against changes in 2157 the interpreter code. It also enables test stubs for eBPF maps and 2158 verifier used by user space verifier testsuite. 2159 2160 If unsure, say N. 2161 2162config TEST_BLACKHOLE_DEV 2163 tristate "Test blackhole netdev functionality" 2164 depends on m && NET 2165 help 2166 This builds the "test_blackhole_dev" module that validates the 2167 data path through this blackhole netdev. 2168 2169 If unsure, say N. 2170 2171config FIND_BIT_BENCHMARK 2172 tristate "Test find_bit functions" 2173 help 2174 This builds the "test_find_bit" module that measure find_*_bit() 2175 functions performance. 2176 2177 If unsure, say N. 2178 2179config TEST_FIRMWARE 2180 tristate "Test firmware loading via userspace interface" 2181 depends on FW_LOADER 2182 help 2183 This builds the "test_firmware" module that creates a userspace 2184 interface for testing firmware loading. This can be used to 2185 control the triggering of firmware loading without needing an 2186 actual firmware-using device. The contents can be rechecked by 2187 userspace. 2188 2189 If unsure, say N. 2190 2191config TEST_SYSCTL 2192 tristate "sysctl test driver" 2193 depends on PROC_SYSCTL 2194 help 2195 This builds the "test_sysctl" module. This driver enables to test the 2196 proc sysctl interfaces available to drivers safely without affecting 2197 production knobs which might alter system functionality. 2198 2199 If unsure, say N. 2200 2201config SYSCTL_KUNIT_TEST 2202 tristate "KUnit test for sysctl" if !KUNIT_ALL_TESTS 2203 depends on KUNIT 2204 default KUNIT_ALL_TESTS 2205 help 2206 This builds the proc sysctl unit test, which runs on boot. 2207 Tests the API contract and implementation correctness of sysctl. 2208 For more information on KUnit and unit tests in general please refer 2209 to the KUnit documentation in Documentation/dev-tools/kunit/. 2210 2211 If unsure, say N. 2212 2213config LIST_KUNIT_TEST 2214 tristate "KUnit Test for Kernel Linked-list structures" if !KUNIT_ALL_TESTS 2215 depends on KUNIT 2216 default KUNIT_ALL_TESTS 2217 help 2218 This builds the linked list KUnit test suite. 2219 It tests that the API and basic functionality of the list_head type 2220 and associated macros. 2221 2222 KUnit tests run during boot and output the results to the debug log 2223 in TAP format (https://testanything.org/). Only useful for kernel devs 2224 running the KUnit test harness, and not intended for inclusion into a 2225 production build. 2226 2227 For more information on KUnit and unit tests in general please refer 2228 to the KUnit documentation in Documentation/dev-tools/kunit/. 2229 2230 If unsure, say N. 2231 2232config LINEAR_RANGES_TEST 2233 tristate "KUnit test for linear_ranges" 2234 depends on KUNIT 2235 select LINEAR_RANGES 2236 help 2237 This builds the linear_ranges unit test, which runs on boot. 2238 Tests the linear_ranges logic correctness. 2239 For more information on KUnit and unit tests in general please refer 2240 to the KUnit documentation in Documentation/dev-tools/kunit/. 2241 2242 If unsure, say N. 2243 2244config BITS_TEST 2245 tristate "KUnit test for bits.h" 2246 depends on KUNIT 2247 help 2248 This builds the bits unit test. 2249 Tests the logic of macros defined in bits.h. 2250 For more information on KUnit and unit tests in general please refer 2251 to the KUnit documentation in Documentation/dev-tools/kunit/. 2252 2253 If unsure, say N. 2254 2255config TEST_UDELAY 2256 tristate "udelay test driver" 2257 help 2258 This builds the "udelay_test" module that helps to make sure 2259 that udelay() is working properly. 2260 2261 If unsure, say N. 2262 2263config TEST_STATIC_KEYS 2264 tristate "Test static keys" 2265 depends on m 2266 help 2267 Test the static key interfaces. 2268 2269 If unsure, say N. 2270 2271config TEST_KMOD 2272 tristate "kmod stress tester" 2273 depends on m 2274 depends on NETDEVICES && NET_CORE && INET # for TUN 2275 depends on BLOCK 2276 select TEST_LKM 2277 select XFS_FS 2278 select TUN 2279 select BTRFS_FS 2280 help 2281 Test the kernel's module loading mechanism: kmod. kmod implements 2282 support to load modules using the Linux kernel's usermode helper. 2283 This test provides a series of tests against kmod. 2284 2285 Although technically you can either build test_kmod as a module or 2286 into the kernel we disallow building it into the kernel since 2287 it stress tests request_module() and this will very likely cause 2288 some issues by taking over precious threads available from other 2289 module load requests, ultimately this could be fatal. 2290 2291 To run tests run: 2292 2293 tools/testing/selftests/kmod/kmod.sh --help 2294 2295 If unsure, say N. 2296 2297config TEST_DEBUG_VIRTUAL 2298 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 2299 depends on DEBUG_VIRTUAL 2300 help 2301 Test the kernel's ability to detect incorrect calls to 2302 virt_to_phys() done against the non-linear part of the 2303 kernel's virtual address map. 2304 2305 If unsure, say N. 2306 2307config TEST_MEMCAT_P 2308 tristate "Test memcat_p() helper function" 2309 help 2310 Test the memcat_p() helper for correctly merging two 2311 pointer arrays together. 2312 2313 If unsure, say N. 2314 2315config TEST_LIVEPATCH 2316 tristate "Test livepatching" 2317 default n 2318 depends on DYNAMIC_DEBUG 2319 depends on LIVEPATCH 2320 depends on m 2321 help 2322 Test kernel livepatching features for correctness. The tests will 2323 load test modules that will be livepatched in various scenarios. 2324 2325 To run all the livepatching tests: 2326 2327 make -C tools/testing/selftests TARGETS=livepatch run_tests 2328 2329 Alternatively, individual tests may be invoked: 2330 2331 tools/testing/selftests/livepatch/test-callbacks.sh 2332 tools/testing/selftests/livepatch/test-livepatch.sh 2333 tools/testing/selftests/livepatch/test-shadow-vars.sh 2334 2335 If unsure, say N. 2336 2337config TEST_OBJAGG 2338 tristate "Perform selftest on object aggreration manager" 2339 default n 2340 depends on OBJAGG 2341 help 2342 Enable this option to test object aggregation manager on boot 2343 (or module load). 2344 2345 2346config TEST_STACKINIT 2347 tristate "Test level of stack variable initialization" 2348 help 2349 Test if the kernel is zero-initializing stack variables and 2350 padding. Coverage is controlled by compiler flags, 2351 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF, 2352 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. 2353 2354 If unsure, say N. 2355 2356config TEST_MEMINIT 2357 tristate "Test heap/page initialization" 2358 help 2359 Test if the kernel is zero-initializing heap and page allocations. 2360 This can be useful to test init_on_alloc and init_on_free features. 2361 2362 If unsure, say N. 2363 2364config TEST_HMM 2365 tristate "Test HMM (Heterogeneous Memory Management)" 2366 depends on TRANSPARENT_HUGEPAGE 2367 depends on DEVICE_PRIVATE 2368 select HMM_MIRROR 2369 select MMU_NOTIFIER 2370 help 2371 This is a pseudo device driver solely for testing HMM. 2372 Say M here if you want to build the HMM test module. 2373 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. 2374 2375 If unsure, say N. 2376 2377config TEST_FREE_PAGES 2378 tristate "Test freeing pages" 2379 help 2380 Test that a memory leak does not occur due to a race between 2381 freeing a block of pages and a speculative page reference. 2382 Loading this module is safe if your kernel has the bug fixed. 2383 If the bug is not fixed, it will leak gigabytes of memory and 2384 probably OOM your system. 2385 2386config TEST_FPU 2387 tristate "Test floating point operations in kernel space" 2388 depends on X86 && !KCOV_INSTRUMENT_ALL 2389 help 2390 Enable this option to add /sys/kernel/debug/selftest_helpers/test_fpu 2391 which will trigger a sequence of floating point operations. This is used 2392 for self-testing floating point control register setting in 2393 kernel_fpu_begin(). 2394 2395 If unsure, say N. 2396 2397endif # RUNTIME_TESTING_MENU 2398 2399config MEMTEST 2400 bool "Memtest" 2401 help 2402 This option adds a kernel parameter 'memtest', which allows memtest 2403 to be set. 2404 memtest=0, mean disabled; -- default 2405 memtest=1, mean do 1 test pattern; 2406 ... 2407 memtest=17, mean do 17 test patterns. 2408 If you are unsure how to answer this question, answer N. 2409 2410 2411 2412config HYPERV_TESTING 2413 bool "Microsoft Hyper-V driver testing" 2414 default n 2415 depends on HYPERV && DEBUG_FS 2416 help 2417 Select this option to enable Hyper-V vmbus testing. 2418 2419endmenu # "Kernel Testing and Coverage" 2420 2421endmenu # Kernel hacking 2422