1ec8f24b7SThomas Gleixner# SPDX-License-Identifier: GPL-2.0-only 2face4374SRoman Zippelconfig DEFCONFIG_LIST 3face4374SRoman Zippel string 4b2670eacSPaolo 'Blaisorblade' Giarrusso depends on !UML 5face4374SRoman Zippel option defconfig_list 647f38ae0SRob Landley default "/lib/modules/$(shell,uname -r)/.config" 7face4374SRoman Zippel default "/etc/kernel-config" 847f38ae0SRob Landley default "/boot/config-$(shell,uname -r)" 92a86f661SMasahiro Yamada default "arch/$(SRCARCH)/configs/$(KBUILD_DEFCONFIG)" 10face4374SRoman Zippel 118b59cd81SMasahiro Yamadaconfig CC_VERSION_TEXT 128b59cd81SMasahiro Yamada string 138b59cd81SMasahiro Yamada default "$(CC_VERSION_TEXT)" 148b59cd81SMasahiro Yamada help 158b59cd81SMasahiro Yamada This is used in unclear ways: 168b59cd81SMasahiro Yamada 178b59cd81SMasahiro Yamada - Re-run Kconfig when the compiler is updated 188b59cd81SMasahiro Yamada The 'default' property references the environment variable, 198b59cd81SMasahiro Yamada CC_VERSION_TEXT so it is recorded in include/config/auto.conf.cmd. 208b59cd81SMasahiro Yamada When the compiler is updated, Kconfig will be invoked. 218b59cd81SMasahiro Yamada 228b59cd81SMasahiro Yamada - Ensure full rebuild when the compier is updated 238b59cd81SMasahiro Yamada include/linux/kconfig.h contains this option in the comment line so 248b59cd81SMasahiro Yamada fixdep adds include/config/cc/version/text.h into the auto-generated 258b59cd81SMasahiro Yamada dependency. When the compiler is updated, syncconfig will touch it 268b59cd81SMasahiro Yamada and then every file will be rebuilt. 278b59cd81SMasahiro Yamada 28a4353898SMasahiro Yamadaconfig CC_IS_GCC 29e33ae3edSMasahiro Yamada def_bool $(success,echo "$(CC_VERSION_TEXT)" | grep -q gcc) 30a4353898SMasahiro Yamada 31a4353898SMasahiro Yamadaconfig GCC_VERSION 32a4353898SMasahiro Yamada int 33fa7295abSMasahiro Yamada default $(shell,$(srctree)/scripts/gcc-version.sh $(CC)) if CC_IS_GCC 34a4353898SMasahiro Yamada default 0 35a4353898SMasahiro Yamada 369553d16fSAmit Daniel Kachhapconfig LD_VERSION 379553d16fSAmit Daniel Kachhap int 389553d16fSAmit Daniel Kachhap default $(shell,$(LD) --version | $(srctree)/scripts/ld-version.sh) 399553d16fSAmit Daniel Kachhap 40469cb737SMasahiro Yamadaconfig CC_IS_CLANG 41e33ae3edSMasahiro Yamada def_bool $(success,echo "$(CC_VERSION_TEXT)" | grep -q clang) 42469cb737SMasahiro Yamada 43b744b43fSSami Tolvanenconfig LD_IS_LLD 44b744b43fSSami Tolvanen def_bool $(success,$(LD) -v | head -n 1 | grep -q LLD) 45b744b43fSSami Tolvanen 46469cb737SMasahiro Yamadaconfig CLANG_VERSION 47469cb737SMasahiro Yamada int 48469cb737SMasahiro Yamada default $(shell,$(srctree)/scripts/clang-version.sh $(CC)) 49469cb737SMasahiro Yamada 501a927fd3SMasahiro Yamadaconfig CC_CAN_LINK 519371f86eSMasahiro Yamada bool 529371f86eSMasahiro Yamada default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(m64-flag)) if 64BIT 539371f86eSMasahiro Yamada default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(m32-flag)) 541a927fd3SMasahiro Yamada 55b1183b6dSMasahiro Yamadaconfig CC_CAN_LINK_STATIC 56b1183b6dSMasahiro Yamada bool 57b1183b6dSMasahiro Yamada default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) -static $(m64-flag)) if 64BIT 58b1183b6dSMasahiro Yamada default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) -static $(m32-flag)) 59c65eacbeSAndy Lutomirski 60e9666d10SMasahiro Yamadaconfig CC_HAS_ASM_GOTO 61e9666d10SMasahiro Yamada def_bool $(success,$(srctree)/scripts/gcc-goto.sh $(CC)) 62e9666d10SMasahiro Yamada 63587f1701SNick Desaulniersconfig CC_HAS_ASM_GOTO_OUTPUT 64587f1701SNick Desaulniers depends on CC_HAS_ASM_GOTO 65587f1701SNick Desaulniers def_bool $(success,echo 'int foo(int x) { asm goto ("": "=r"(x) ::: bar); return x; bar: return 0; }' | $(CC) -x c - -c -o /dev/null) 66587f1701SNick Desaulniers 675cf896fbSPeter Collingbourneconfig TOOLS_SUPPORT_RELR 682d122942SWill Deacon def_bool $(success,env "CC=$(CC)" "LD=$(LD)" "NM=$(NM)" "OBJCOPY=$(OBJCOPY)" $(srctree)/scripts/tools-support-relr.sh) 695cf896fbSPeter Collingbourne 70eb111869SRasmus Villemoesconfig CC_HAS_ASM_INLINE 71eb111869SRasmus Villemoes def_bool $(success,echo 'void foo(void) { asm inline (""); }' | $(CC) -x c - -c -o /dev/null) 72eb111869SRasmus Villemoes 73b99b87f7SPeter Oberparleiterconfig CONSTRUCTORS 74b99b87f7SPeter Oberparleiter bool 7587c9366eSJohannes Berg depends on !UML 76b99b87f7SPeter Oberparleiter 77e360adbeSPeter Zijlstraconfig IRQ_WORK 78e360adbeSPeter Zijlstra bool 79e360adbeSPeter Zijlstra 8010916706SShile Zhangconfig BUILDTIME_TABLE_SORT 811dbdc6f1SDavid Daney bool 821dbdc6f1SDavid Daney 83c65eacbeSAndy Lutomirskiconfig THREAD_INFO_IN_TASK 84c65eacbeSAndy Lutomirski bool 85c65eacbeSAndy Lutomirski help 86c65eacbeSAndy Lutomirski Select this to move thread_info off the stack into task_struct. To 87c65eacbeSAndy Lutomirski make this work, an arch will need to remove all thread_info fields 88c65eacbeSAndy Lutomirski except flags and fix any runtime bugs. 89c65eacbeSAndy Lutomirski 90c6c314a6SAndy Lutomirski One subtle change that will be needed is to use try_get_task_stack() 91c6c314a6SAndy Lutomirski and put_task_stack() in save_thread_stack_tsk() and get_wchan(). 92c6c314a6SAndy Lutomirski 93ff0cfc66SAl Boldimenu "General setup" 941da177e4SLinus Torvalds 951da177e4SLinus Torvaldsconfig BROKEN 961da177e4SLinus Torvalds bool 971da177e4SLinus Torvalds 981da177e4SLinus Torvaldsconfig BROKEN_ON_SMP 991da177e4SLinus Torvalds bool 1001da177e4SLinus Torvalds depends on BROKEN || !SMP 1011da177e4SLinus Torvalds default y 1021da177e4SLinus Torvalds 1031da177e4SLinus Torvaldsconfig INIT_ENV_ARG_LIMIT 1041da177e4SLinus Torvalds int 105dd673bcaSAdrian Bunk default 32 if !UML 106dd673bcaSAdrian Bunk default 128 if UML 1071da177e4SLinus Torvalds help 10834ad92c2SRandy Dunlap Maximum of each of the number of arguments and environment 10934ad92c2SRandy Dunlap variables passed to init from the kernel command line. 1101da177e4SLinus Torvalds 1114bb16672SJiri Slabyconfig COMPILE_TEST 1124bb16672SJiri Slaby bool "Compile also drivers which will not load" 113bc083a64SRichard Weinberger depends on !UML 1144bb16672SJiri Slaby default n 1154bb16672SJiri Slaby help 1164bb16672SJiri Slaby Some drivers can be compiled on a different platform than they are 1174bb16672SJiri Slaby intended to be run on. Despite they cannot be loaded there (or even 1184bb16672SJiri Slaby when they load they cannot be used due to missing HW support), 1194bb16672SJiri Slaby developers still, opposing to distributors, might want to build such 1204bb16672SJiri Slaby drivers to compile-test them. 1214bb16672SJiri Slaby 1224bb16672SJiri Slaby If you are a developer and want to build everything available, say Y 1234bb16672SJiri Slaby here. If you are a user/distributor, say N here to exclude useless 1244bb16672SJiri Slaby drivers to be distributed. 1254bb16672SJiri Slaby 126d6fc9fcbSMasahiro Yamadaconfig UAPI_HEADER_TEST 127d6fc9fcbSMasahiro Yamada bool "Compile test UAPI headers" 128fcbb8461SMasahiro Yamada depends on HEADERS_INSTALL && CC_CAN_LINK 129d6fc9fcbSMasahiro Yamada help 130d6fc9fcbSMasahiro Yamada Compile test headers exported to user-space to ensure they are 131d6fc9fcbSMasahiro Yamada self-contained, i.e. compilable as standalone units. 132d6fc9fcbSMasahiro Yamada 133d6fc9fcbSMasahiro Yamada If you are a developer or tester and want to ensure the exported 134d6fc9fcbSMasahiro Yamada headers are self-contained, say Y here. Otherwise, choose N. 135d6fc9fcbSMasahiro Yamada 1361da177e4SLinus Torvaldsconfig LOCALVERSION 1371da177e4SLinus Torvalds string "Local version - append to kernel release" 1381da177e4SLinus Torvalds help 1391da177e4SLinus Torvalds Append an extra string to the end of your kernel version. 1401da177e4SLinus Torvalds This will show up when you type uname, for example. 1411da177e4SLinus Torvalds The string you set here will be appended after the contents of 1421da177e4SLinus Torvalds any files with a filename matching localversion* in your 1431da177e4SLinus Torvalds object and source tree, in that order. Your total string can 1441da177e4SLinus Torvalds be a maximum of 64 characters. 1451da177e4SLinus Torvalds 146aaebf433SRyan Andersonconfig LOCALVERSION_AUTO 147aaebf433SRyan Anderson bool "Automatically append version information to the version string" 148aaebf433SRyan Anderson default y 149ac3339baSAlexey Dobriyan depends on !COMPILE_TEST 150aaebf433SRyan Anderson help 151aaebf433SRyan Anderson This will try to automatically determine if the current tree is a 1526e5a5420SRobert P. J. Day release tree by looking for git tags that belong to the current 1536e5a5420SRobert P. J. Day top of tree revision. 154aaebf433SRyan Anderson 155aaebf433SRyan Anderson A string of the format -gxxxxxxxx will be added to the localversion 1566e5a5420SRobert P. J. Day if a git-based tree is found. The string generated by this will be 157aaebf433SRyan Anderson appended after any matching localversion* files, and after the value 1586e5a5420SRobert P. J. Day set in CONFIG_LOCALVERSION. 159aaebf433SRyan Anderson 1606e5a5420SRobert P. J. Day (The actual string used here is the first eight characters produced 1616e5a5420SRobert P. J. Day by running the command: 1626e5a5420SRobert P. J. Day 1636e5a5420SRobert P. J. Day $ git rev-parse --verify HEAD 1646e5a5420SRobert P. J. Day 1656e5a5420SRobert P. J. Day which is done within the script "scripts/setlocalversion".) 166aaebf433SRyan Anderson 1679afb719eSLaura Abbottconfig BUILD_SALT 1689afb719eSLaura Abbott string "Build ID Salt" 1699afb719eSLaura Abbott default "" 1709afb719eSLaura Abbott help 1719afb719eSLaura Abbott The build ID is used to link binaries and their debug info. Setting 1729afb719eSLaura Abbott this option will use the value in the calculation of the build id. 1739afb719eSLaura Abbott This is mostly useful for distributions which want to ensure the 1749afb719eSLaura Abbott build is unique between builds. It's safe to leave the default. 1759afb719eSLaura Abbott 1762e9f3bddSH. Peter Anvinconfig HAVE_KERNEL_GZIP 1772e9f3bddSH. Peter Anvin bool 1782e9f3bddSH. Peter Anvin 1792e9f3bddSH. Peter Anvinconfig HAVE_KERNEL_BZIP2 1802e9f3bddSH. Peter Anvin bool 1812e9f3bddSH. Peter Anvin 1822e9f3bddSH. Peter Anvinconfig HAVE_KERNEL_LZMA 1832e9f3bddSH. Peter Anvin bool 1842e9f3bddSH. Peter Anvin 1853ebe1243SLasse Collinconfig HAVE_KERNEL_XZ 1863ebe1243SLasse Collin bool 1873ebe1243SLasse Collin 1887dd65febSAlbin Tonnerreconfig HAVE_KERNEL_LZO 1897dd65febSAlbin Tonnerre bool 1907dd65febSAlbin Tonnerre 191e76e1fdfSKyungsik Leeconfig HAVE_KERNEL_LZ4 192e76e1fdfSKyungsik Lee bool 193e76e1fdfSKyungsik Lee 194f16466afSVasily Gorbikconfig HAVE_KERNEL_UNCOMPRESSED 195f16466afSVasily Gorbik bool 196f16466afSVasily Gorbik 19730d65dbfSAlain Knaffchoice 19830d65dbfSAlain Knaff prompt "Kernel compression mode" 19930d65dbfSAlain Knaff default KERNEL_GZIP 200f16466afSVasily Gorbik depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_UNCOMPRESSED 20130d65dbfSAlain Knaff help 20230d65dbfSAlain Knaff The linux kernel is a kind of self-extracting executable. 20330d65dbfSAlain Knaff Several compression algorithms are available, which differ 20430d65dbfSAlain Knaff in efficiency, compression and decompression speed. 20530d65dbfSAlain Knaff Compression speed is only relevant when building a kernel. 20630d65dbfSAlain Knaff Decompression speed is relevant at each boot. 20730d65dbfSAlain Knaff 20830d65dbfSAlain Knaff If you have any problems with bzip2 or lzma compressed 20930d65dbfSAlain Knaff kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 21030d65dbfSAlain Knaff version of this functionality (bzip2 only), for 2.4, was 21130d65dbfSAlain Knaff supplied by Christian Ludwig) 21230d65dbfSAlain Knaff 21330d65dbfSAlain Knaff High compression options are mostly useful for users, who 21430d65dbfSAlain Knaff are low on disk space (embedded systems), but for whom ram 21530d65dbfSAlain Knaff size matters less. 21630d65dbfSAlain Knaff 21730d65dbfSAlain Knaff If in doubt, select 'gzip' 21830d65dbfSAlain Knaff 21930d65dbfSAlain Knaffconfig KERNEL_GZIP 22030d65dbfSAlain Knaff bool "Gzip" 2212e9f3bddSH. Peter Anvin depends on HAVE_KERNEL_GZIP 22230d65dbfSAlain Knaff help 2237dd65febSAlbin Tonnerre The old and tried gzip compression. It provides a good balance 2247dd65febSAlbin Tonnerre between compression ratio and decompression speed. 22530d65dbfSAlain Knaff 22630d65dbfSAlain Knaffconfig KERNEL_BZIP2 22730d65dbfSAlain Knaff bool "Bzip2" 2282e9f3bddSH. Peter Anvin depends on HAVE_KERNEL_BZIP2 22930d65dbfSAlain Knaff help 23030d65dbfSAlain Knaff Its compression ratio and speed is intermediate. 2310a4dd35cSRandy Dunlap Decompression speed is slowest among the choices. The kernel 2322e9f3bddSH. Peter Anvin size is about 10% smaller with bzip2, in comparison to gzip. 2332e9f3bddSH. Peter Anvin Bzip2 uses a large amount of memory. For modern kernels you 2342e9f3bddSH. Peter Anvin will need at least 8MB RAM or more for booting. 23530d65dbfSAlain Knaff 23630d65dbfSAlain Knaffconfig KERNEL_LZMA 23730d65dbfSAlain Knaff bool "LZMA" 2382e9f3bddSH. Peter Anvin depends on HAVE_KERNEL_LZMA 23930d65dbfSAlain Knaff help 2400a4dd35cSRandy Dunlap This compression algorithm's ratio is best. Decompression speed 2410a4dd35cSRandy Dunlap is between gzip and bzip2. Compression is slowest. 2420a4dd35cSRandy Dunlap The kernel size is about 33% smaller with LZMA in comparison to gzip. 24330d65dbfSAlain Knaff 2443ebe1243SLasse Collinconfig KERNEL_XZ 2453ebe1243SLasse Collin bool "XZ" 2463ebe1243SLasse Collin depends on HAVE_KERNEL_XZ 2473ebe1243SLasse Collin help 2483ebe1243SLasse Collin XZ uses the LZMA2 algorithm and instruction set specific 2493ebe1243SLasse Collin BCJ filters which can improve compression ratio of executable 2503ebe1243SLasse Collin code. The size of the kernel is about 30% smaller with XZ in 2513ebe1243SLasse Collin comparison to gzip. On architectures for which there is a BCJ 2523ebe1243SLasse Collin filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 2533ebe1243SLasse Collin will create a few percent smaller kernel than plain LZMA. 2543ebe1243SLasse Collin 2553ebe1243SLasse Collin The speed is about the same as with LZMA: The decompression 2563ebe1243SLasse Collin speed of XZ is better than that of bzip2 but worse than gzip 2573ebe1243SLasse Collin and LZO. Compression is slow. 2583ebe1243SLasse Collin 2597dd65febSAlbin Tonnerreconfig KERNEL_LZO 2607dd65febSAlbin Tonnerre bool "LZO" 2617dd65febSAlbin Tonnerre depends on HAVE_KERNEL_LZO 2627dd65febSAlbin Tonnerre help 2630a4dd35cSRandy Dunlap Its compression ratio is the poorest among the choices. The kernel 264681b3049SStephan Sperber size is about 10% bigger than gzip; however its speed 2657dd65febSAlbin Tonnerre (both compression and decompression) is the fastest. 2667dd65febSAlbin Tonnerre 267e76e1fdfSKyungsik Leeconfig KERNEL_LZ4 268e76e1fdfSKyungsik Lee bool "LZ4" 269e76e1fdfSKyungsik Lee depends on HAVE_KERNEL_LZ4 270e76e1fdfSKyungsik Lee help 271e76e1fdfSKyungsik Lee LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding. 272e76e1fdfSKyungsik Lee A preliminary version of LZ4 de/compression tool is available at 273e76e1fdfSKyungsik Lee <https://code.google.com/p/lz4/>. 274e76e1fdfSKyungsik Lee 275e76e1fdfSKyungsik Lee Its compression ratio is worse than LZO. The size of the kernel 276e76e1fdfSKyungsik Lee is about 8% bigger than LZO. But the decompression speed is 277e76e1fdfSKyungsik Lee faster than LZO. 278e76e1fdfSKyungsik Lee 279f16466afSVasily Gorbikconfig KERNEL_UNCOMPRESSED 280f16466afSVasily Gorbik bool "None" 281f16466afSVasily Gorbik depends on HAVE_KERNEL_UNCOMPRESSED 282f16466afSVasily Gorbik help 283f16466afSVasily Gorbik Produce uncompressed kernel image. This option is usually not what 284f16466afSVasily Gorbik you want. It is useful for debugging the kernel in slow simulation 285f16466afSVasily Gorbik environments, where decompressing and moving the kernel is awfully 286f16466afSVasily Gorbik slow. This option allows early boot code to skip the decompressor 287f16466afSVasily Gorbik and jump right at uncompressed kernel image. 288f16466afSVasily Gorbik 28930d65dbfSAlain Knaffendchoice 29030d65dbfSAlain Knaff 291ada4ab7aSChris Downconfig DEFAULT_INIT 292ada4ab7aSChris Down string "Default init path" 293ada4ab7aSChris Down default "" 294ada4ab7aSChris Down help 295ada4ab7aSChris Down This option determines the default init for the system if no init= 296ada4ab7aSChris Down option is passed on the kernel command line. If the requested path is 297ada4ab7aSChris Down not present, we will still then move on to attempting further 298ada4ab7aSChris Down locations (e.g. /sbin/init, etc). If this is empty, we will just use 299ada4ab7aSChris Down the fallback list when init= is not passed. 300ada4ab7aSChris Down 301bd5dc17bSJosh Triplettconfig DEFAULT_HOSTNAME 302bd5dc17bSJosh Triplett string "Default hostname" 303bd5dc17bSJosh Triplett default "(none)" 304bd5dc17bSJosh Triplett help 305bd5dc17bSJosh Triplett This option determines the default system hostname before userspace 306bd5dc17bSJosh Triplett calls sethostname(2). The kernel traditionally uses "(none)" here, 307bd5dc17bSJosh Triplett but you may wish to use a different default here to make a minimal 308bd5dc17bSJosh Triplett system more usable with less configuration. 309bd5dc17bSJosh Triplett 31017c46a6aSChristoph Hellwig# 31117c46a6aSChristoph Hellwig# For some reason microblaze and nios2 hard code SWAP=n. Hopefully we can 31217c46a6aSChristoph Hellwig# add proper SWAP support to them, in which case this can be remove. 31317c46a6aSChristoph Hellwig# 31417c46a6aSChristoph Hellwigconfig ARCH_NO_SWAP 31517c46a6aSChristoph Hellwig bool 31617c46a6aSChristoph Hellwig 3171da177e4SLinus Torvaldsconfig SWAP 3181da177e4SLinus Torvalds bool "Support for paging of anonymous memory (swap)" 31917c46a6aSChristoph Hellwig depends on MMU && BLOCK && !ARCH_NO_SWAP 3201da177e4SLinus Torvalds default y 3211da177e4SLinus Torvalds help 3221da177e4SLinus Torvalds This option allows you to choose whether you want to have support 3231da177e4SLinus Torvalds for so called swap devices or swap files in your kernel that are 3241da177e4SLinus Torvalds used to provide more virtual memory than the actual RAM present 3251da177e4SLinus Torvalds in your computer. If unsure say Y. 3261da177e4SLinus Torvalds 3271da177e4SLinus Torvaldsconfig SYSVIPC 3281da177e4SLinus Torvalds bool "System V IPC" 329a7f7f624SMasahiro Yamada help 3301da177e4SLinus Torvalds Inter Process Communication is a suite of library functions and 3311da177e4SLinus Torvalds system calls which let processes (running programs) synchronize and 3321da177e4SLinus Torvalds exchange information. It is generally considered to be a good thing, 3331da177e4SLinus Torvalds and some programs won't run unless you say Y here. In particular, if 3341da177e4SLinus Torvalds you want to run the DOS emulator dosemu under Linux (read the 3351da177e4SLinus Torvalds DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 3361da177e4SLinus Torvalds you'll need to say Y here. 3371da177e4SLinus Torvalds 3381da177e4SLinus Torvalds You can find documentation about IPC with "info ipc" and also in 3391da177e4SLinus Torvalds section 6.4 of the Linux Programmer's Guide, available from 3401da177e4SLinus Torvalds <http://www.tldp.org/guides.html>. 3411da177e4SLinus Torvalds 342a5494dcdSEric W. Biedermanconfig SYSVIPC_SYSCTL 343a5494dcdSEric W. Biederman bool 344a5494dcdSEric W. Biederman depends on SYSVIPC 345a5494dcdSEric W. Biederman depends on SYSCTL 346a5494dcdSEric W. Biederman default y 347a5494dcdSEric W. Biederman 3481da177e4SLinus Torvaldsconfig POSIX_MQUEUE 3491da177e4SLinus Torvalds bool "POSIX Message Queues" 35019c92399SKees Cook depends on NET 351a7f7f624SMasahiro Yamada help 3521da177e4SLinus Torvalds POSIX variant of message queues is a part of IPC. In POSIX message 3531da177e4SLinus Torvalds queues every message has a priority which decides about succession 3541da177e4SLinus Torvalds of receiving it by a process. If you want to compile and run 3551da177e4SLinus Torvalds programs written e.g. for Solaris with use of its POSIX message 356b0e37650SRobert P. J. Day queues (functions mq_*) say Y here. 3571da177e4SLinus Torvalds 3581da177e4SLinus Torvalds POSIX message queues are visible as a filesystem called 'mqueue' 3591da177e4SLinus Torvalds and can be mounted somewhere if you want to do filesystem 3601da177e4SLinus Torvalds operations on message queues. 3611da177e4SLinus Torvalds 3621da177e4SLinus Torvalds If unsure, say Y. 3631da177e4SLinus Torvalds 364bdc8e5f8SSerge E. Hallynconfig POSIX_MQUEUE_SYSCTL 365bdc8e5f8SSerge E. Hallyn bool 366bdc8e5f8SSerge E. Hallyn depends on POSIX_MQUEUE 367bdc8e5f8SSerge E. Hallyn depends on SYSCTL 368bdc8e5f8SSerge E. Hallyn default y 369bdc8e5f8SSerge E. Hallyn 370c73be61cSDavid Howellsconfig WATCH_QUEUE 371c73be61cSDavid Howells bool "General notification queue" 372c73be61cSDavid Howells default n 373c73be61cSDavid Howells help 374c73be61cSDavid Howells 375c73be61cSDavid Howells This is a general notification queue for the kernel to pass events to 376c73be61cSDavid Howells userspace by splicing them into pipes. It can be used in conjunction 377c73be61cSDavid Howells with watches for key/keyring change notifications and device 378c73be61cSDavid Howells notifications. 379c73be61cSDavid Howells 380c73be61cSDavid Howells See Documentation/watch_queue.rst 381c73be61cSDavid Howells 382226b4ccdSKonstantin Khlebnikovconfig CROSS_MEMORY_ATTACH 383226b4ccdSKonstantin Khlebnikov bool "Enable process_vm_readv/writev syscalls" 384226b4ccdSKonstantin Khlebnikov depends on MMU 385226b4ccdSKonstantin Khlebnikov default y 386226b4ccdSKonstantin Khlebnikov help 387226b4ccdSKonstantin Khlebnikov Enabling this option adds the system calls process_vm_readv and 388226b4ccdSKonstantin Khlebnikov process_vm_writev which allow a process with the correct privileges 389a2a368d9SGeert Uytterhoeven to directly read from or write to another process' address space. 390226b4ccdSKonstantin Khlebnikov See the man page for more details. 391226b4ccdSKonstantin Khlebnikov 39269369a70SJosh Triplettconfig USELIB 39369369a70SJosh Triplett bool "uselib syscall" 394b2113a41SRiku Voipio def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION 39569369a70SJosh Triplett help 39669369a70SJosh Triplett This option enables the uselib syscall, a system call used in the 39769369a70SJosh Triplett dynamic linker from libc5 and earlier. glibc does not use this 39869369a70SJosh Triplett system call. If you intend to run programs built on libc5 or 39969369a70SJosh Triplett earlier, you may need to enable this syscall. Current systems 40069369a70SJosh Triplett running glibc can safely disable this. 40169369a70SJosh Triplett 4021da177e4SLinus Torvaldsconfig AUDIT 4031da177e4SLinus Torvalds bool "Auditing support" 404804a6a49SChris Wright depends on NET 4051da177e4SLinus Torvalds help 4061da177e4SLinus Torvalds Enable auditing infrastructure that can be used with another 4071da177e4SLinus Torvalds kernel subsystem, such as SELinux (which requires this for 408cb74ed27SPaul Moore logging of avc messages output). System call auditing is included 409cb74ed27SPaul Moore on architectures which support it. 4101da177e4SLinus Torvalds 4117a017721SAKASHI Takahiroconfig HAVE_ARCH_AUDITSYSCALL 4127a017721SAKASHI Takahiro bool 4137a017721SAKASHI Takahiro 4141da177e4SLinus Torvaldsconfig AUDITSYSCALL 415cb74ed27SPaul Moore def_bool y 4167a017721SAKASHI Takahiro depends on AUDIT && HAVE_ARCH_AUDITSYSCALL 41728a3a7ebSEric Paris select FSNOTIFY 41874c3cbe3SAl Viro 419d9817ebeSThomas Gleixnersource "kernel/irq/Kconfig" 420764e0da1SThomas Gleixnersource "kernel/time/Kconfig" 42187a4c375SChristoph Hellwigsource "kernel/Kconfig.preempt" 422d9817ebeSThomas Gleixner 423391dc69cSFrederic Weisbeckermenu "CPU/Task time and stats accounting" 424391dc69cSFrederic Weisbecker 425abf917cdSFrederic Weisbeckerconfig VIRT_CPU_ACCOUNTING 426abf917cdSFrederic Weisbecker bool 427abf917cdSFrederic Weisbecker 428fdf9c356SFrederic Weisbeckerchoice 429fdf9c356SFrederic Weisbecker prompt "Cputime accounting" 430fdf9c356SFrederic Weisbecker default TICK_CPU_ACCOUNTING if !PPC64 43102fc8d37SStephen Rothwell default VIRT_CPU_ACCOUNTING_NATIVE if PPC64 432fdf9c356SFrederic Weisbecker 433fdf9c356SFrederic Weisbecker# Kind of a stub config for the pure tick based cputime accounting 434fdf9c356SFrederic Weisbeckerconfig TICK_CPU_ACCOUNTING 435fdf9c356SFrederic Weisbecker bool "Simple tick based cputime accounting" 436c58b0df1SFrederic Weisbecker depends on !S390 && !NO_HZ_FULL 437fdf9c356SFrederic Weisbecker help 438fdf9c356SFrederic Weisbecker This is the basic tick based cputime accounting that maintains 439fdf9c356SFrederic Weisbecker statistics about user, system and idle time spent on per jiffies 440fdf9c356SFrederic Weisbecker granularity. 441fdf9c356SFrederic Weisbecker 442fdf9c356SFrederic Weisbecker If unsure, say Y. 443fdf9c356SFrederic Weisbecker 444abf917cdSFrederic Weisbeckerconfig VIRT_CPU_ACCOUNTING_NATIVE 445391dc69cSFrederic Weisbecker bool "Deterministic task and CPU time accounting" 446c58b0df1SFrederic Weisbecker depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL 447abf917cdSFrederic Weisbecker select VIRT_CPU_ACCOUNTING 448391dc69cSFrederic Weisbecker help 449391dc69cSFrederic Weisbecker Select this option to enable more accurate task and CPU time 450391dc69cSFrederic Weisbecker accounting. This is done by reading a CPU counter on each 451391dc69cSFrederic Weisbecker kernel entry and exit and on transitions within the kernel 452391dc69cSFrederic Weisbecker between system, softirq and hardirq state, so there is a 453391dc69cSFrederic Weisbecker small performance impact. In the case of s390 or IBM POWER > 5, 454391dc69cSFrederic Weisbecker this also enables accounting of stolen time on logically-partitioned 455391dc69cSFrederic Weisbecker systems. 456391dc69cSFrederic Weisbecker 457abf917cdSFrederic Weisbeckerconfig VIRT_CPU_ACCOUNTING_GEN 458abf917cdSFrederic Weisbecker bool "Full dynticks CPU time accounting" 459ff3fb254SKevin Hilman depends on HAVE_CONTEXT_TRACKING 460554b0004SKevin Hilman depends on HAVE_VIRT_CPU_ACCOUNTING_GEN 461041a1574SArnd Bergmann depends on GENERIC_CLOCKEVENTS 462abf917cdSFrederic Weisbecker select VIRT_CPU_ACCOUNTING 463abf917cdSFrederic Weisbecker select CONTEXT_TRACKING 464abf917cdSFrederic Weisbecker help 465abf917cdSFrederic Weisbecker Select this option to enable task and CPU time accounting on full 466abf917cdSFrederic Weisbecker dynticks systems. This accounting is implemented by watching every 467abf917cdSFrederic Weisbecker kernel-user boundaries using the context tracking subsystem. 468abf917cdSFrederic Weisbecker The accounting is thus performed at the expense of some significant 469abf917cdSFrederic Weisbecker overhead. 470abf917cdSFrederic Weisbecker 471abf917cdSFrederic Weisbecker For now this is only useful if you are working on the full 472abf917cdSFrederic Weisbecker dynticks subsystem development. 473abf917cdSFrederic Weisbecker 474abf917cdSFrederic Weisbecker If unsure, say N. 475abf917cdSFrederic Weisbecker 476b58c3584SRik van Rielendchoice 477b58c3584SRik van Riel 478fdf9c356SFrederic Weisbeckerconfig IRQ_TIME_ACCOUNTING 479fdf9c356SFrederic Weisbecker bool "Fine granularity task level IRQ time accounting" 480b58c3584SRik van Riel depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE 481fdf9c356SFrederic Weisbecker help 482fdf9c356SFrederic Weisbecker Select this option to enable fine granularity task irq time 483fdf9c356SFrederic Weisbecker accounting. This is done by reading a timestamp on each 484fdf9c356SFrederic Weisbecker transitions between softirq and hardirq state, so there can be a 485fdf9c356SFrederic Weisbecker small performance impact. 486fdf9c356SFrederic Weisbecker 487fdf9c356SFrederic Weisbecker If in doubt, say N here. 488fdf9c356SFrederic Weisbecker 48911d4afd4SVincent Guittotconfig HAVE_SCHED_AVG_IRQ 49011d4afd4SVincent Guittot def_bool y 49111d4afd4SVincent Guittot depends on IRQ_TIME_ACCOUNTING || PARAVIRT_TIME_ACCOUNTING 49211d4afd4SVincent Guittot depends on SMP 49311d4afd4SVincent Guittot 49476504793SThara Gopinathconfig SCHED_THERMAL_PRESSURE 49576504793SThara Gopinath bool "Enable periodic averaging of thermal pressure" 49676504793SThara Gopinath depends on SMP 49776504793SThara Gopinath 498391dc69cSFrederic Weisbeckerconfig BSD_PROCESS_ACCT 499391dc69cSFrederic Weisbecker bool "BSD Process Accounting" 5002813893fSIulia Manda depends on MULTIUSER 501391dc69cSFrederic Weisbecker help 502391dc69cSFrederic Weisbecker If you say Y here, a user level program will be able to instruct the 503391dc69cSFrederic Weisbecker kernel (via a special system call) to write process accounting 504391dc69cSFrederic Weisbecker information to a file: whenever a process exits, information about 505391dc69cSFrederic Weisbecker that process will be appended to the file by the kernel. The 506391dc69cSFrederic Weisbecker information includes things such as creation time, owning user, 507391dc69cSFrederic Weisbecker command name, memory usage, controlling terminal etc. (the complete 508391dc69cSFrederic Weisbecker list is in the struct acct in <file:include/linux/acct.h>). It is 509391dc69cSFrederic Weisbecker up to the user level program to do useful things with this 510391dc69cSFrederic Weisbecker information. This is generally a good idea, so say Y. 511391dc69cSFrederic Weisbecker 512391dc69cSFrederic Weisbeckerconfig BSD_PROCESS_ACCT_V3 513391dc69cSFrederic Weisbecker bool "BSD Process Accounting version 3 file format" 514391dc69cSFrederic Weisbecker depends on BSD_PROCESS_ACCT 515391dc69cSFrederic Weisbecker default n 516391dc69cSFrederic Weisbecker help 517391dc69cSFrederic Weisbecker If you say Y here, the process accounting information is written 518391dc69cSFrederic Weisbecker in a new file format that also logs the process IDs of each 5193903bf94SRandy Dunlap process and its parent. Note that this file format is incompatible 520391dc69cSFrederic Weisbecker with previous v0/v1/v2 file formats, so you will need updated tools 521391dc69cSFrederic Weisbecker for processing it. A preliminary version of these tools is available 522391dc69cSFrederic Weisbecker at <http://www.gnu.org/software/acct/>. 523391dc69cSFrederic Weisbecker 524391dc69cSFrederic Weisbeckerconfig TASKSTATS 52519c92399SKees Cook bool "Export task/process statistics through netlink" 526391dc69cSFrederic Weisbecker depends on NET 5272813893fSIulia Manda depends on MULTIUSER 528391dc69cSFrederic Weisbecker default n 529391dc69cSFrederic Weisbecker help 530391dc69cSFrederic Weisbecker Export selected statistics for tasks/processes through the 531391dc69cSFrederic Weisbecker generic netlink interface. Unlike BSD process accounting, the 532391dc69cSFrederic Weisbecker statistics are available during the lifetime of tasks/processes as 533391dc69cSFrederic Weisbecker responses to commands. Like BSD accounting, they are sent to user 534391dc69cSFrederic Weisbecker space on task exit. 535391dc69cSFrederic Weisbecker 536391dc69cSFrederic Weisbecker Say N if unsure. 537391dc69cSFrederic Weisbecker 538391dc69cSFrederic Weisbeckerconfig TASK_DELAY_ACCT 53919c92399SKees Cook bool "Enable per-task delay accounting" 540391dc69cSFrederic Weisbecker depends on TASKSTATS 541f6db8347SNaveen N. Rao select SCHED_INFO 542391dc69cSFrederic Weisbecker help 543391dc69cSFrederic Weisbecker Collect information on time spent by a task waiting for system 544391dc69cSFrederic Weisbecker resources like cpu, synchronous block I/O completion and swapping 545391dc69cSFrederic Weisbecker in pages. Such statistics can help in setting a task's priorities 546391dc69cSFrederic Weisbecker relative to other tasks for cpu, io, rss limits etc. 547391dc69cSFrederic Weisbecker 548391dc69cSFrederic Weisbecker Say N if unsure. 549391dc69cSFrederic Weisbecker 550391dc69cSFrederic Weisbeckerconfig TASK_XACCT 55119c92399SKees Cook bool "Enable extended accounting over taskstats" 552391dc69cSFrederic Weisbecker depends on TASKSTATS 553391dc69cSFrederic Weisbecker help 554391dc69cSFrederic Weisbecker Collect extended task accounting data and send the data 555391dc69cSFrederic Weisbecker to userland for processing over the taskstats interface. 556391dc69cSFrederic Weisbecker 557391dc69cSFrederic Weisbecker Say N if unsure. 558391dc69cSFrederic Weisbecker 559391dc69cSFrederic Weisbeckerconfig TASK_IO_ACCOUNTING 56019c92399SKees Cook bool "Enable per-task storage I/O accounting" 561391dc69cSFrederic Weisbecker depends on TASK_XACCT 562391dc69cSFrederic Weisbecker help 563391dc69cSFrederic Weisbecker Collect information on the number of bytes of storage I/O which this 564391dc69cSFrederic Weisbecker task has caused. 565391dc69cSFrederic Weisbecker 566391dc69cSFrederic Weisbecker Say N if unsure. 567391dc69cSFrederic Weisbecker 568eb414681SJohannes Weinerconfig PSI 569eb414681SJohannes Weiner bool "Pressure stall information tracking" 570eb414681SJohannes Weiner help 571eb414681SJohannes Weiner Collect metrics that indicate how overcommitted the CPU, memory, 572eb414681SJohannes Weiner and IO capacity are in the system. 573eb414681SJohannes Weiner 574eb414681SJohannes Weiner If you say Y here, the kernel will create /proc/pressure/ with the 575eb414681SJohannes Weiner pressure statistics files cpu, memory, and io. These will indicate 576eb414681SJohannes Weiner the share of walltime in which some or all tasks in the system are 577eb414681SJohannes Weiner delayed due to contention of the respective resource. 578eb414681SJohannes Weiner 5792ce7135aSJohannes Weiner In kernels with cgroup support, cgroups (cgroup2 only) will 5802ce7135aSJohannes Weiner have cpu.pressure, memory.pressure, and io.pressure files, 5812ce7135aSJohannes Weiner which aggregate pressure stalls for the grouped tasks only. 5822ce7135aSJohannes Weiner 583c3123552SMauro Carvalho Chehab For more details see Documentation/accounting/psi.rst. 584eb414681SJohannes Weiner 585eb414681SJohannes Weiner Say N if unsure. 586eb414681SJohannes Weiner 587e0c27447SJohannes Weinerconfig PSI_DEFAULT_DISABLED 588e0c27447SJohannes Weiner bool "Require boot parameter to enable pressure stall information tracking" 589e0c27447SJohannes Weiner default n 590e0c27447SJohannes Weiner depends on PSI 591e0c27447SJohannes Weiner help 592e0c27447SJohannes Weiner If set, pressure stall information tracking will be disabled 593428a1cb4SBaruch Siach per default but can be enabled through passing psi=1 on the 594428a1cb4SBaruch Siach kernel commandline during boot. 595e0c27447SJohannes Weiner 5967b2489d3SJohannes Weiner This feature adds some code to the task wakeup and sleep 5977b2489d3SJohannes Weiner paths of the scheduler. The overhead is too low to affect 5987b2489d3SJohannes Weiner common scheduling-intense workloads in practice (such as 5997b2489d3SJohannes Weiner webservers, memcache), but it does show up in artificial 6007b2489d3SJohannes Weiner scheduler stress tests, such as hackbench. 6017b2489d3SJohannes Weiner 6027b2489d3SJohannes Weiner If you are paranoid and not sure what the kernel will be 6037b2489d3SJohannes Weiner used for, say Y. 6047b2489d3SJohannes Weiner 6057b2489d3SJohannes Weiner Say N if unsure. 6067b2489d3SJohannes Weiner 607391dc69cSFrederic Weisbeckerendmenu # "CPU/Task time and stats accounting" 608391dc69cSFrederic Weisbecker 6095c4991e2SFrederic Weisbeckerconfig CPU_ISOLATION 6105c4991e2SFrederic Weisbecker bool "CPU isolation" 611414a2dc1SGeert Uytterhoeven depends on SMP || COMPILE_TEST 6122c43838cSFrederic Weisbecker default y 6135c4991e2SFrederic Weisbecker help 6145c4991e2SFrederic Weisbecker Make sure that CPUs running critical tasks are not disturbed by 6155c4991e2SFrederic Weisbecker any source of "noise" such as unbound workqueues, timers, kthreads... 6162c43838cSFrederic Weisbecker Unbound jobs get offloaded to housekeeping CPUs. This is driven by 6172c43838cSFrederic Weisbecker the "isolcpus=" boot parameter. 6182c43838cSFrederic Weisbecker 6192c43838cSFrederic Weisbecker Say Y if unsure. 6205c4991e2SFrederic Weisbecker 6210af92d46SPaul E. McKenneysource "kernel/rcu/Kconfig" 622c903ff83SMike Travis 623de5b56baSVivek Goyalconfig BUILD_BIN2C 624de5b56baSVivek Goyal bool 625de5b56baSVivek Goyal default n 626de5b56baSVivek Goyal 6271da177e4SLinus Torvaldsconfig IKCONFIG 628f2443ab6SRoss Biro tristate "Kernel .config support" 629a7f7f624SMasahiro Yamada help 6301da177e4SLinus Torvalds This option enables the complete Linux kernel ".config" file 6311da177e4SLinus Torvalds contents to be saved in the kernel. It provides documentation 6321da177e4SLinus Torvalds of which kernel options are used in a running kernel or in an 6331da177e4SLinus Torvalds on-disk kernel. This information can be extracted from the kernel 6341da177e4SLinus Torvalds image file with the script scripts/extract-ikconfig and used as 6351da177e4SLinus Torvalds input to rebuild the current kernel or to build another kernel. 6361da177e4SLinus Torvalds It can also be extracted from a running kernel by reading 6371da177e4SLinus Torvalds /proc/config.gz if enabled (below). 6381da177e4SLinus Torvalds 6391da177e4SLinus Torvaldsconfig IKCONFIG_PROC 6401da177e4SLinus Torvalds bool "Enable access to .config through /proc/config.gz" 6411da177e4SLinus Torvalds depends on IKCONFIG && PROC_FS 642a7f7f624SMasahiro Yamada help 6431da177e4SLinus Torvalds This option enables access to the kernel configuration file 6441da177e4SLinus Torvalds through /proc/config.gz. 6451da177e4SLinus Torvalds 646f7b101d3SJoel Fernandes (Google)config IKHEADERS 647f7b101d3SJoel Fernandes (Google) tristate "Enable kernel headers through /sys/kernel/kheaders.tar.xz" 648f7b101d3SJoel Fernandes (Google) depends on SYSFS 64943d8ce9dSJoel Fernandes (Google) help 650f7b101d3SJoel Fernandes (Google) This option enables access to the in-kernel headers that are generated during 651f7b101d3SJoel Fernandes (Google) the build process. These can be used to build eBPF tracing programs, 652f7b101d3SJoel Fernandes (Google) or similar programs. If you build the headers as a module, a module called 653f7b101d3SJoel Fernandes (Google) kheaders.ko is built which can be loaded on-demand to get access to headers. 65443d8ce9dSJoel Fernandes (Google) 655794543a2SAlistair John Strachanconfig LOG_BUF_SHIFT 656794543a2SAlistair John Strachan int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 657*550c10d2SJohn Ogness range 12 25 if !H8300 658*550c10d2SJohn Ogness range 12 19 if H8300 659f17a32e9SAdrian Bunk default 17 660361e9dfbSJosh Triplett depends on PRINTK 661794543a2SAlistair John Strachan help 66223b2899fSLuis R. Rodriguez Select the minimal kernel log buffer size as a power of 2. 66323b2899fSLuis R. Rodriguez The final size is affected by LOG_CPU_MAX_BUF_SHIFT config 66423b2899fSLuis R. Rodriguez parameter, see below. Any higher size also might be forced 66523b2899fSLuis R. Rodriguez by "log_buf_len" boot parameter. 66623b2899fSLuis R. Rodriguez 667f17a32e9SAdrian Bunk Examples: 668f17a32e9SAdrian Bunk 17 => 128 KB 669f17a32e9SAdrian Bunk 16 => 64 KB 670f17a32e9SAdrian Bunk 15 => 32 KB 671f17a32e9SAdrian Bunk 14 => 16 KB 672794543a2SAlistair John Strachan 13 => 8 KB 673794543a2SAlistair John Strachan 12 => 4 KB 674794543a2SAlistair John Strachan 67523b2899fSLuis R. Rodriguezconfig LOG_CPU_MAX_BUF_SHIFT 67623b2899fSLuis R. Rodriguez int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)" 6772240a31dSGeert Uytterhoeven depends on SMP 67823b2899fSLuis R. Rodriguez range 0 21 67923b2899fSLuis R. Rodriguez default 12 if !BASE_SMALL 68023b2899fSLuis R. Rodriguez default 0 if BASE_SMALL 681361e9dfbSJosh Triplett depends on PRINTK 68223b2899fSLuis R. Rodriguez help 68323b2899fSLuis R. Rodriguez This option allows to increase the default ring buffer size 68423b2899fSLuis R. Rodriguez according to the number of CPUs. The value defines the contribution 68523b2899fSLuis R. Rodriguez of each CPU as a power of 2. The used space is typically only few 68623b2899fSLuis R. Rodriguez lines however it might be much more when problems are reported, 68723b2899fSLuis R. Rodriguez e.g. backtraces. 68823b2899fSLuis R. Rodriguez 68923b2899fSLuis R. Rodriguez The increased size means that a new buffer has to be allocated and 69023b2899fSLuis R. Rodriguez the original static one is unused. It makes sense only on systems 69123b2899fSLuis R. Rodriguez with more CPUs. Therefore this value is used only when the sum of 69223b2899fSLuis R. Rodriguez contributions is greater than the half of the default kernel ring 69323b2899fSLuis R. Rodriguez buffer as defined by LOG_BUF_SHIFT. The default values are set 69423b2899fSLuis R. Rodriguez so that more than 64 CPUs are needed to trigger the allocation. 69523b2899fSLuis R. Rodriguez 69623b2899fSLuis R. Rodriguez Also this option is ignored when "log_buf_len" kernel parameter is 69723b2899fSLuis R. Rodriguez used as it forces an exact (power of two) size of the ring buffer. 69823b2899fSLuis R. Rodriguez 69923b2899fSLuis R. Rodriguez The number of possible CPUs is used for this computation ignoring 7005e0d8d59SGeert Uytterhoeven hotplugging making the computation optimal for the worst case 7015e0d8d59SGeert Uytterhoeven scenario while allowing a simple algorithm to be used from bootup. 70223b2899fSLuis R. Rodriguez 70323b2899fSLuis R. Rodriguez Examples shift values and their meaning: 70423b2899fSLuis R. Rodriguez 17 => 128 KB for each CPU 70523b2899fSLuis R. Rodriguez 16 => 64 KB for each CPU 70623b2899fSLuis R. Rodriguez 15 => 32 KB for each CPU 70723b2899fSLuis R. Rodriguez 14 => 16 KB for each CPU 70823b2899fSLuis R. Rodriguez 13 => 8 KB for each CPU 70923b2899fSLuis R. Rodriguez 12 => 4 KB for each CPU 71023b2899fSLuis R. Rodriguez 711f92bac3bSSergey Senozhatskyconfig PRINTK_SAFE_LOG_BUF_SHIFT 712f92bac3bSSergey Senozhatsky int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)" 713427934b8SPetr Mladek range 10 21 714427934b8SPetr Mladek default 13 715f92bac3bSSergey Senozhatsky depends on PRINTK 716427934b8SPetr Mladek help 717f92bac3bSSergey Senozhatsky Select the size of an alternate printk per-CPU buffer where messages 718f92bac3bSSergey Senozhatsky printed from usafe contexts are temporary stored. One example would 719f92bac3bSSergey Senozhatsky be NMI messages, another one - printk recursion. The messages are 720f92bac3bSSergey Senozhatsky copied to the main log buffer in a safe context to avoid a deadlock. 721f92bac3bSSergey Senozhatsky The value defines the size as a power of 2. 722427934b8SPetr Mladek 723f92bac3bSSergey Senozhatsky Those messages are rare and limited. The largest one is when 724427934b8SPetr Mladek a backtrace is printed. It usually fits into 4KB. Select 725427934b8SPetr Mladek 8KB if you want to be on the safe side. 726427934b8SPetr Mladek 727427934b8SPetr Mladek Examples: 728427934b8SPetr Mladek 17 => 128 KB for each CPU 729427934b8SPetr Mladek 16 => 64 KB for each CPU 730427934b8SPetr Mladek 15 => 32 KB for each CPU 731427934b8SPetr Mladek 14 => 16 KB for each CPU 732427934b8SPetr Mladek 13 => 8 KB for each CPU 733427934b8SPetr Mladek 12 => 4 KB for each CPU 734427934b8SPetr Mladek 7355cdc38f9SKAMEZAWA Hiroyuki# 7365cdc38f9SKAMEZAWA Hiroyuki# Architectures with an unreliable sched_clock() should select this: 7375cdc38f9SKAMEZAWA Hiroyuki# 7385cdc38f9SKAMEZAWA Hiroyukiconfig HAVE_UNSTABLE_SCHED_CLOCK 7395cdc38f9SKAMEZAWA Hiroyuki bool 7405cdc38f9SKAMEZAWA Hiroyuki 74138ff87f7SStephen Boydconfig GENERIC_SCHED_CLOCK 74238ff87f7SStephen Boyd bool 74338ff87f7SStephen Boyd 74469842cbaSPatrick Bellasimenu "Scheduler features" 74569842cbaSPatrick Bellasi 74669842cbaSPatrick Bellasiconfig UCLAMP_TASK 74769842cbaSPatrick Bellasi bool "Enable utilization clamping for RT/FAIR tasks" 74869842cbaSPatrick Bellasi depends on CPU_FREQ_GOV_SCHEDUTIL 74969842cbaSPatrick Bellasi help 75069842cbaSPatrick Bellasi This feature enables the scheduler to track the clamped utilization 75169842cbaSPatrick Bellasi of each CPU based on RUNNABLE tasks scheduled on that CPU. 75269842cbaSPatrick Bellasi 75369842cbaSPatrick Bellasi With this option, the user can specify the min and max CPU 75469842cbaSPatrick Bellasi utilization allowed for RUNNABLE tasks. The max utilization defines 75569842cbaSPatrick Bellasi the maximum frequency a task should use while the min utilization 75669842cbaSPatrick Bellasi defines the minimum frequency it should use. 75769842cbaSPatrick Bellasi 75869842cbaSPatrick Bellasi Both min and max utilization clamp values are hints to the scheduler, 75969842cbaSPatrick Bellasi aiming at improving its frequency selection policy, but they do not 76069842cbaSPatrick Bellasi enforce or grant any specific bandwidth for tasks. 76169842cbaSPatrick Bellasi 76269842cbaSPatrick Bellasi If in doubt, say N. 76369842cbaSPatrick Bellasi 76469842cbaSPatrick Bellasiconfig UCLAMP_BUCKETS_COUNT 76569842cbaSPatrick Bellasi int "Number of supported utilization clamp buckets" 76669842cbaSPatrick Bellasi range 5 20 76769842cbaSPatrick Bellasi default 5 76869842cbaSPatrick Bellasi depends on UCLAMP_TASK 76969842cbaSPatrick Bellasi help 77069842cbaSPatrick Bellasi Defines the number of clamp buckets to use. The range of each bucket 77169842cbaSPatrick Bellasi will be SCHED_CAPACITY_SCALE/UCLAMP_BUCKETS_COUNT. The higher the 77269842cbaSPatrick Bellasi number of clamp buckets the finer their granularity and the higher 77369842cbaSPatrick Bellasi the precision of clamping aggregation and tracking at run-time. 77469842cbaSPatrick Bellasi 77569842cbaSPatrick Bellasi For example, with the minimum configuration value we will have 5 77669842cbaSPatrick Bellasi clamp buckets tracking 20% utilization each. A 25% boosted tasks will 77769842cbaSPatrick Bellasi be refcounted in the [20..39]% bucket and will set the bucket clamp 77869842cbaSPatrick Bellasi effective value to 25%. 77969842cbaSPatrick Bellasi If a second 30% boosted task should be co-scheduled on the same CPU, 78069842cbaSPatrick Bellasi that task will be refcounted in the same bucket of the first task and 78169842cbaSPatrick Bellasi it will boost the bucket clamp effective value to 30%. 78269842cbaSPatrick Bellasi The clamp effective value of a bucket is reset to its nominal value 78369842cbaSPatrick Bellasi (20% in the example above) when there are no more tasks refcounted in 78469842cbaSPatrick Bellasi that bucket. 78569842cbaSPatrick Bellasi 78669842cbaSPatrick Bellasi An additional boost/capping margin can be added to some tasks. In the 78769842cbaSPatrick Bellasi example above the 25% task will be boosted to 30% until it exits the 78869842cbaSPatrick Bellasi CPU. If that should be considered not acceptable on certain systems, 78969842cbaSPatrick Bellasi it's always possible to reduce the margin by increasing the number of 79069842cbaSPatrick Bellasi clamp buckets to trade off used memory for run-time tracking 79169842cbaSPatrick Bellasi precision. 79269842cbaSPatrick Bellasi 79369842cbaSPatrick Bellasi If in doubt, use the default value. 79469842cbaSPatrick Bellasi 79569842cbaSPatrick Bellasiendmenu 79669842cbaSPatrick Bellasi 797be3a7284SAndrea Arcangeli# 798be3a7284SAndrea Arcangeli# For architectures that want to enable the support for NUMA-affine scheduler 799be3a7284SAndrea Arcangeli# balancing logic: 800be3a7284SAndrea Arcangeli# 801be3a7284SAndrea Arcangeliconfig ARCH_SUPPORTS_NUMA_BALANCING 802be3a7284SAndrea Arcangeli bool 803be3a7284SAndrea Arcangeli 804be5e610cSPeter Zijlstra# 80572b252aeSMel Gorman# For architectures that prefer to flush all TLBs after a number of pages 80672b252aeSMel Gorman# are unmapped instead of sending one IPI per page to flush. The architecture 80772b252aeSMel Gorman# must provide guarantees on what happens if a clean TLB cache entry is 80872b252aeSMel Gorman# written after the unmap. Details are in mm/rmap.c near the check for 80972b252aeSMel Gorman# should_defer_flush. The architecture should also consider if the full flush 81072b252aeSMel Gorman# and the refill costs are offset by the savings of sending fewer IPIs. 81172b252aeSMel Gormanconfig ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH 81272b252aeSMel Gorman bool 81372b252aeSMel Gorman 814c12d3362SArd Biesheuvelconfig CC_HAS_INT128 8153a7c7331SMasahiro Yamada def_bool !$(cc-option,$(m64-flag) -D__SIZEOF_INT128__=0) && 64BIT 816c12d3362SArd Biesheuvel 81772b252aeSMel Gorman# 818be5e610cSPeter Zijlstra# For architectures that know their GCC __int128 support is sound 819be5e610cSPeter Zijlstra# 820be5e610cSPeter Zijlstraconfig ARCH_SUPPORTS_INT128 821be5e610cSPeter Zijlstra bool 822be5e610cSPeter Zijlstra 823be3a7284SAndrea Arcangeli# For architectures that (ab)use NUMA to represent different memory regions 824be3a7284SAndrea Arcangeli# all cpu-local but of different latencies, such as SuperH. 825be3a7284SAndrea Arcangeli# 826be3a7284SAndrea Arcangeliconfig ARCH_WANT_NUMA_VARIABLE_LOCALITY 827be3a7284SAndrea Arcangeli bool 828be3a7284SAndrea Arcangeli 829be3a7284SAndrea Arcangeliconfig NUMA_BALANCING 830be3a7284SAndrea Arcangeli bool "Memory placement aware NUMA scheduler" 831be3a7284SAndrea Arcangeli depends on ARCH_SUPPORTS_NUMA_BALANCING 832be3a7284SAndrea Arcangeli depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY 833be3a7284SAndrea Arcangeli depends on SMP && NUMA && MIGRATION 834be3a7284SAndrea Arcangeli help 835be3a7284SAndrea Arcangeli This option adds support for automatic NUMA aware memory/task placement. 836be3a7284SAndrea Arcangeli The mechanism is quite primitive and is based on migrating memory when 8376d56a410SPaul Gortmaker it has references to the node the task is running on. 838be3a7284SAndrea Arcangeli 839be3a7284SAndrea Arcangeli This system will be inactive on UMA systems. 840be3a7284SAndrea Arcangeli 8416f7c97e8SAneesh Kumar K.Vconfig NUMA_BALANCING_DEFAULT_ENABLED 8426f7c97e8SAneesh Kumar K.V bool "Automatically enable NUMA aware memory/task placement" 8436f7c97e8SAneesh Kumar K.V default y 8446f7c97e8SAneesh Kumar K.V depends on NUMA_BALANCING 8456f7c97e8SAneesh Kumar K.V help 8466f7c97e8SAneesh Kumar K.V If set, automatic NUMA balancing will be enabled if running on a NUMA 8476f7c97e8SAneesh Kumar K.V machine. 8486f7c97e8SAneesh Kumar K.V 84923964d2dSLi Zefanmenuconfig CGROUPS 8506341e62bSChristoph Jaeger bool "Control Group support" 8512bd59d48STejun Heo select KERNFS 852ddbcc7e8SPaul Menage help 85323964d2dSLi Zefan This option adds support for grouping sets of processes together, for 8545cdc38f9SKAMEZAWA Hiroyuki use with process control subsystems such as Cpusets, CFS, memory 8555cdc38f9SKAMEZAWA Hiroyuki controls or device isolation. 8565cdc38f9SKAMEZAWA Hiroyuki See 857d6a3b247SMauro Carvalho Chehab - Documentation/scheduler/sched-design-CFS.rst (CFS) 858da82c92fSMauro Carvalho Chehab - Documentation/admin-guide/cgroup-v1/ (features for grouping, isolation 85945ce80fbSLi Zefan and resource control) 860ddbcc7e8SPaul Menage 861ddbcc7e8SPaul Menage Say N if unsure. 862ddbcc7e8SPaul Menage 86323964d2dSLi Zefanif CGROUPS 86423964d2dSLi Zefan 8653e32cb2eSJohannes Weinerconfig PAGE_COUNTER 8663e32cb2eSJohannes Weiner bool 8673e32cb2eSJohannes Weiner 868c255a458SAndrew Mortonconfig MEMCG 869a0166ec4SJohannes Weiner bool "Memory controller" 8703e32cb2eSJohannes Weiner select PAGE_COUNTER 87179bd9814STejun Heo select EVENTFD 87200f0b825SBalbir Singh help 873a0166ec4SJohannes Weiner Provides control over the memory footprint of tasks in a cgroup. 87400f0b825SBalbir Singh 875c255a458SAndrew Mortonconfig MEMCG_SWAP 8762d1c4980SJohannes Weiner bool 877c255a458SAndrew Morton depends on MEMCG && SWAP 878a42c390cSMichal Hocko default y 879c077719bSKAMEZAWA Hiroyuki 88084c07d11SKirill Tkhaiconfig MEMCG_KMEM 88184c07d11SKirill Tkhai bool 88284c07d11SKirill Tkhai depends on MEMCG && !SLOB 88384c07d11SKirill Tkhai default y 88484c07d11SKirill Tkhai 8856bf024e6SJohannes Weinerconfig BLK_CGROUP 8866bf024e6SJohannes Weiner bool "IO controller" 8876bf024e6SJohannes Weiner depends on BLOCK 8882bc64a20SAneesh Kumar K.V default n 889a7f7f624SMasahiro Yamada help 8906bf024e6SJohannes Weiner Generic block IO controller cgroup interface. This is the common 8916bf024e6SJohannes Weiner cgroup interface which should be used by various IO controlling 8926bf024e6SJohannes Weiner policies. 8932bc64a20SAneesh Kumar K.V 8946bf024e6SJohannes Weiner Currently, CFQ IO scheduler uses it to recognize task groups and 8956bf024e6SJohannes Weiner control disk bandwidth allocation (proportional time slice allocation) 8966bf024e6SJohannes Weiner to such task groups. It is also used by bio throttling logic in 8976bf024e6SJohannes Weiner block layer to implement upper limit in IO rates on a device. 898e5d1367fSStephane Eranian 8996bf024e6SJohannes Weiner This option only enables generic Block IO controller infrastructure. 9006bf024e6SJohannes Weiner One needs to also enable actual IO controlling logic/policy. For 9016bf024e6SJohannes Weiner enabling proportional weight division of disk bandwidth in CFQ, set 9027baf2199SKrzysztof Kozlowski CONFIG_BFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 9036bf024e6SJohannes Weiner CONFIG_BLK_DEV_THROTTLING=y. 9046bf024e6SJohannes Weiner 905da82c92fSMauro Carvalho Chehab See Documentation/admin-guide/cgroup-v1/blkio-controller.rst for more information. 9066bf024e6SJohannes Weiner 9076bf024e6SJohannes Weinerconfig CGROUP_WRITEBACK 9086bf024e6SJohannes Weiner bool 9096bf024e6SJohannes Weiner depends on MEMCG && BLK_CGROUP 9106bf024e6SJohannes Weiner default y 911e5d1367fSStephane Eranian 9127c941438SDhaval Gianimenuconfig CGROUP_SCHED 913a0166ec4SJohannes Weiner bool "CPU controller" 9147c941438SDhaval Giani default n 9157c941438SDhaval Giani help 9167c941438SDhaval Giani This feature lets CPU scheduler recognize task groups and control CPU 9177c941438SDhaval Giani bandwidth allocation to such task groups. It uses cgroups to group 9187c941438SDhaval Giani tasks. 9197c941438SDhaval Giani 9207c941438SDhaval Gianiif CGROUP_SCHED 9217c941438SDhaval Gianiconfig FAIR_GROUP_SCHED 9227c941438SDhaval Giani bool "Group scheduling for SCHED_OTHER" 9237c941438SDhaval Giani depends on CGROUP_SCHED 9247c941438SDhaval Giani default CGROUP_SCHED 9257c941438SDhaval Giani 926ab84d31eSPaul Turnerconfig CFS_BANDWIDTH 927ab84d31eSPaul Turner bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 928ab84d31eSPaul Turner depends on FAIR_GROUP_SCHED 929ab84d31eSPaul Turner default n 930ab84d31eSPaul Turner help 931ab84d31eSPaul Turner This option allows users to define CPU bandwidth rates (limits) for 932ab84d31eSPaul Turner tasks running within the fair group scheduler. Groups with no limit 933ab84d31eSPaul Turner set are considered to be unconstrained and will run with no 934ab84d31eSPaul Turner restriction. 935d6a3b247SMauro Carvalho Chehab See Documentation/scheduler/sched-bwc.rst for more information. 936ab84d31eSPaul Turner 9377c941438SDhaval Gianiconfig RT_GROUP_SCHED 9387c941438SDhaval Giani bool "Group scheduling for SCHED_RR/FIFO" 9397c941438SDhaval Giani depends on CGROUP_SCHED 9407c941438SDhaval Giani default n 9417c941438SDhaval Giani help 9427c941438SDhaval Giani This feature lets you explicitly allocate real CPU bandwidth 94332bd7eb5SLi Zefan to task groups. If enabled, it will also make it impossible to 9447c941438SDhaval Giani schedule realtime tasks for non-root users until you allocate 9457c941438SDhaval Giani realtime bandwidth for them. 946d6a3b247SMauro Carvalho Chehab See Documentation/scheduler/sched-rt-group.rst for more information. 9477c941438SDhaval Giani 9487c941438SDhaval Gianiendif #CGROUP_SCHED 9497c941438SDhaval Giani 9502480c093SPatrick Bellasiconfig UCLAMP_TASK_GROUP 9512480c093SPatrick Bellasi bool "Utilization clamping per group of tasks" 9522480c093SPatrick Bellasi depends on CGROUP_SCHED 9532480c093SPatrick Bellasi depends on UCLAMP_TASK 9542480c093SPatrick Bellasi default n 9552480c093SPatrick Bellasi help 9562480c093SPatrick Bellasi This feature enables the scheduler to track the clamped utilization 9572480c093SPatrick Bellasi of each CPU based on RUNNABLE tasks currently scheduled on that CPU. 9582480c093SPatrick Bellasi 9592480c093SPatrick Bellasi When this option is enabled, the user can specify a min and max 9602480c093SPatrick Bellasi CPU bandwidth which is allowed for each single task in a group. 9612480c093SPatrick Bellasi The max bandwidth allows to clamp the maximum frequency a task 9622480c093SPatrick Bellasi can use, while the min bandwidth allows to define a minimum 9632480c093SPatrick Bellasi frequency a task will always use. 9642480c093SPatrick Bellasi 9652480c093SPatrick Bellasi When task group based utilization clamping is enabled, an eventually 9662480c093SPatrick Bellasi specified task-specific clamp value is constrained by the cgroup 9672480c093SPatrick Bellasi specified clamp value. Both minimum and maximum task clamping cannot 9682480c093SPatrick Bellasi be bigger than the corresponding clamping defined at task group level. 9692480c093SPatrick Bellasi 9702480c093SPatrick Bellasi If in doubt, say N. 9712480c093SPatrick Bellasi 9726bf024e6SJohannes Weinerconfig CGROUP_PIDS 9736bf024e6SJohannes Weiner bool "PIDs controller" 9746bf024e6SJohannes Weiner help 9756bf024e6SJohannes Weiner Provides enforcement of process number limits in the scope of a 9766bf024e6SJohannes Weiner cgroup. Any attempt to fork more processes than is allowed in the 9776bf024e6SJohannes Weiner cgroup will fail. PIDs are fundamentally a global resource because it 9786bf024e6SJohannes Weiner is fairly trivial to reach PID exhaustion before you reach even a 9796bf024e6SJohannes Weiner conservative kmemcg limit. As a result, it is possible to grind a 9806bf024e6SJohannes Weiner system to halt without being limited by other cgroup policies. The 9816cc578dfSParav Pandit PIDs controller is designed to stop this from happening. 9826bf024e6SJohannes Weiner 9836bf024e6SJohannes Weiner It should be noted that organisational operations (such as attaching 98498076833SJonathan Neuschäfer to a cgroup hierarchy) will *not* be blocked by the PIDs controller, 9856bf024e6SJohannes Weiner since the PIDs limit only affects a process's ability to fork, not to 9866bf024e6SJohannes Weiner attach to a cgroup. 9876bf024e6SJohannes Weiner 98839d3e758SParav Panditconfig CGROUP_RDMA 98939d3e758SParav Pandit bool "RDMA controller" 99039d3e758SParav Pandit help 99139d3e758SParav Pandit Provides enforcement of RDMA resources defined by IB stack. 99239d3e758SParav Pandit It is fairly easy for consumers to exhaust RDMA resources, which 99339d3e758SParav Pandit can result into resource unavailability to other consumers. 99439d3e758SParav Pandit RDMA controller is designed to stop this from happening. 99539d3e758SParav Pandit Attaching processes with active RDMA resources to the cgroup 99639d3e758SParav Pandit hierarchy is allowed even if can cross the hierarchy's limit. 99739d3e758SParav Pandit 9986bf024e6SJohannes Weinerconfig CGROUP_FREEZER 9996bf024e6SJohannes Weiner bool "Freezer controller" 10006bf024e6SJohannes Weiner help 10016bf024e6SJohannes Weiner Provides a way to freeze and unfreeze all tasks in a 10026bf024e6SJohannes Weiner cgroup. 10036bf024e6SJohannes Weiner 1004489c2a20SJohannes Weiner This option affects the ORIGINAL cgroup interface. The cgroup2 memory 1005489c2a20SJohannes Weiner controller includes important in-kernel memory consumers per default. 1006489c2a20SJohannes Weiner 1007489c2a20SJohannes Weiner If you're using cgroup2, say N. 1008489c2a20SJohannes Weiner 10096bf024e6SJohannes Weinerconfig CGROUP_HUGETLB 10106bf024e6SJohannes Weiner bool "HugeTLB controller" 10116bf024e6SJohannes Weiner depends on HUGETLB_PAGE 10126bf024e6SJohannes Weiner select PAGE_COUNTER 1013afc24d49SVivek Goyal default n 10146bf024e6SJohannes Weiner help 10156bf024e6SJohannes Weiner Provides a cgroup controller for HugeTLB pages. 10166bf024e6SJohannes Weiner When you enable this, you can put a per cgroup limit on HugeTLB usage. 10176bf024e6SJohannes Weiner The limit is enforced during page fault. Since HugeTLB doesn't 10186bf024e6SJohannes Weiner support page reclaim, enforcing the limit at page fault time implies 10196bf024e6SJohannes Weiner that, the application will get SIGBUS signal if it tries to access 10206bf024e6SJohannes Weiner HugeTLB pages beyond its limit. This requires the application to know 10216bf024e6SJohannes Weiner beforehand how much HugeTLB pages it would require for its use. The 10226bf024e6SJohannes Weiner control group is tracked in the third page lru pointer. This means 10236bf024e6SJohannes Weiner that we cannot use the controller with huge page less than 3 pages. 1024afc24d49SVivek Goyal 10256bf024e6SJohannes Weinerconfig CPUSETS 10266bf024e6SJohannes Weiner bool "Cpuset controller" 1027e1d4eeecSNicolas Pitre depends on SMP 10286bf024e6SJohannes Weiner help 10296bf024e6SJohannes Weiner This option will let you create and manage CPUSETs which 10306bf024e6SJohannes Weiner allow dynamically partitioning a system into sets of CPUs and 10316bf024e6SJohannes Weiner Memory Nodes and assigning tasks to run only within those sets. 10326bf024e6SJohannes Weiner This is primarily useful on large SMP or NUMA systems. 1033afc24d49SVivek Goyal 10346bf024e6SJohannes Weiner Say N if unsure. 1035afc24d49SVivek Goyal 10366bf024e6SJohannes Weinerconfig PROC_PID_CPUSET 10376bf024e6SJohannes Weiner bool "Include legacy /proc/<pid>/cpuset file" 10386bf024e6SJohannes Weiner depends on CPUSETS 103989e9b9e0STejun Heo default y 104089e9b9e0STejun Heo 10416bf024e6SJohannes Weinerconfig CGROUP_DEVICE 10426bf024e6SJohannes Weiner bool "Device controller" 10436bf024e6SJohannes Weiner help 10446bf024e6SJohannes Weiner Provides a cgroup controller implementing whitelists for 10456bf024e6SJohannes Weiner devices which a process in the cgroup can mknod or open. 10466bf024e6SJohannes Weiner 10476bf024e6SJohannes Weinerconfig CGROUP_CPUACCT 10486bf024e6SJohannes Weiner bool "Simple CPU accounting controller" 10496bf024e6SJohannes Weiner help 10506bf024e6SJohannes Weiner Provides a simple controller for monitoring the 10516bf024e6SJohannes Weiner total CPU consumed by the tasks in a cgroup. 10526bf024e6SJohannes Weiner 10536bf024e6SJohannes Weinerconfig CGROUP_PERF 10546bf024e6SJohannes Weiner bool "Perf controller" 10556bf024e6SJohannes Weiner depends on PERF_EVENTS 10566bf024e6SJohannes Weiner help 10576bf024e6SJohannes Weiner This option extends the perf per-cpu mode to restrict monitoring 10586bf024e6SJohannes Weiner to threads which belong to the cgroup specified and run on the 10596546b19fSNamhyung Kim designated cpu. Or this can be used to have cgroup ID in samples 10606546b19fSNamhyung Kim so that it can monitor performance events among cgroups. 10616bf024e6SJohannes Weiner 10626bf024e6SJohannes Weiner Say N if unsure. 10636bf024e6SJohannes Weiner 106430070984SDaniel Mackconfig CGROUP_BPF 106530070984SDaniel Mack bool "Support for eBPF programs attached to cgroups" 1066483c4933SAndy Lutomirski depends on BPF_SYSCALL 1067483c4933SAndy Lutomirski select SOCK_CGROUP_DATA 106830070984SDaniel Mack help 106930070984SDaniel Mack Allow attaching eBPF programs to a cgroup using the bpf(2) 107030070984SDaniel Mack syscall command BPF_PROG_ATTACH. 107130070984SDaniel Mack 107230070984SDaniel Mack In which context these programs are accessed depends on the type 107330070984SDaniel Mack of attachment. For instance, programs that are attached using 107430070984SDaniel Mack BPF_CGROUP_INET_INGRESS will be executed on the ingress path of 107530070984SDaniel Mack inet sockets. 107630070984SDaniel Mack 10776bf024e6SJohannes Weinerconfig CGROUP_DEBUG 107823b0be48SWaiman Long bool "Debug controller" 10796bf024e6SJohannes Weiner default n 108023b0be48SWaiman Long depends on DEBUG_KERNEL 10816bf024e6SJohannes Weiner help 10826bf024e6SJohannes Weiner This option enables a simple controller that exports 108323b0be48SWaiman Long debugging information about the cgroups framework. This 108423b0be48SWaiman Long controller is for control cgroup debugging only. Its 108523b0be48SWaiman Long interfaces are not stable. 10866bf024e6SJohannes Weiner 10876bf024e6SJohannes Weiner Say N. 10886bf024e6SJohannes Weiner 108973b35147SArnd Bergmannconfig SOCK_CGROUP_DATA 109073b35147SArnd Bergmann bool 109173b35147SArnd Bergmann default n 109273b35147SArnd Bergmann 109323964d2dSLi Zefanendif # CGROUPS 1094c077719bSKAMEZAWA Hiroyuki 10958dd2a82cSDaniel Lezcanomenuconfig NAMESPACES 10966a108a14SDavid Rientjes bool "Namespaces support" if EXPERT 10972813893fSIulia Manda depends on MULTIUSER 10986a108a14SDavid Rientjes default !EXPERT 1099c5289a69SPavel Emelyanov help 1100c5289a69SPavel Emelyanov Provides the way to make tasks work with different objects using 1101c5289a69SPavel Emelyanov the same id. For example same IPC id may refer to different objects 1102c5289a69SPavel Emelyanov or same user id or pid may refer to different tasks when used in 1103c5289a69SPavel Emelyanov different namespaces. 1104c5289a69SPavel Emelyanov 11058dd2a82cSDaniel Lezcanoif NAMESPACES 11068dd2a82cSDaniel Lezcano 110758bfdd6dSPavel Emelyanovconfig UTS_NS 110858bfdd6dSPavel Emelyanov bool "UTS namespace" 110917a6d441SDaniel Lezcano default y 111058bfdd6dSPavel Emelyanov help 111158bfdd6dSPavel Emelyanov In this namespace tasks see different info provided with the 111258bfdd6dSPavel Emelyanov uname() system call 111358bfdd6dSPavel Emelyanov 1114769071acSAndrei Vaginconfig TIME_NS 1115769071acSAndrei Vagin bool "TIME namespace" 1116660fd04fSThomas Gleixner depends on GENERIC_VDSO_TIME_NS 1117769071acSAndrei Vagin default y 1118769071acSAndrei Vagin help 1119769071acSAndrei Vagin In this namespace boottime and monotonic clocks can be set. 1120769071acSAndrei Vagin The time will keep going with the same pace. 1121769071acSAndrei Vagin 1122ae5e1b22SPavel Emelyanovconfig IPC_NS 1123ae5e1b22SPavel Emelyanov bool "IPC namespace" 11248dd2a82cSDaniel Lezcano depends on (SYSVIPC || POSIX_MQUEUE) 112517a6d441SDaniel Lezcano default y 1126ae5e1b22SPavel Emelyanov help 1127ae5e1b22SPavel Emelyanov In this namespace tasks work with IPC ids which correspond to 1128614b84cfSSerge E. Hallyn different IPC objects in different namespaces. 1129ae5e1b22SPavel Emelyanov 1130aee16ce7SPavel Emelyanovconfig USER_NS 113119c92399SKees Cook bool "User namespace" 11325673a94cSEric W. Biederman default n 1133aee16ce7SPavel Emelyanov help 1134aee16ce7SPavel Emelyanov This allows containers, i.e. vservers, to use user namespaces 1135aee16ce7SPavel Emelyanov to provide different user info for different servers. 1136e11f0ae3SEric W. Biederman 1137e11f0ae3SEric W. Biederman When user namespaces are enabled in the kernel it is 1138d886f4e4SJohannes Weiner recommended that the MEMCG option also be enabled and that 1139d886f4e4SJohannes Weiner user-space use the memory control groups to limit the amount 1140d886f4e4SJohannes Weiner of memory a memory unprivileged users can use. 1141e11f0ae3SEric W. Biederman 1142aee16ce7SPavel Emelyanov If unsure, say N. 1143aee16ce7SPavel Emelyanov 114474bd59bbSPavel Emelyanovconfig PID_NS 11459bd38c2cSDaniel Lezcano bool "PID Namespaces" 114617a6d441SDaniel Lezcano default y 114774bd59bbSPavel Emelyanov help 114812d2b8f9SHeikki Orsila Support process id namespaces. This allows having multiple 1149692105b8SMatt LaPlante processes with the same pid as long as they are in different 115074bd59bbSPavel Emelyanov pid namespaces. This is a building block of containers. 115174bd59bbSPavel Emelyanov 1152d6eb633fSMatt Helsleyconfig NET_NS 1153d6eb633fSMatt Helsley bool "Network namespace" 11548dd2a82cSDaniel Lezcano depends on NET 115517a6d441SDaniel Lezcano default y 1156d6eb633fSMatt Helsley help 1157d6eb633fSMatt Helsley Allow user space to create what appear to be multiple instances 1158d6eb633fSMatt Helsley of the network stack. 1159d6eb633fSMatt Helsley 11608dd2a82cSDaniel Lezcanoendif # NAMESPACES 11618dd2a82cSDaniel Lezcano 11625cb366bbSAdrian Reberconfig CHECKPOINT_RESTORE 11635cb366bbSAdrian Reber bool "Checkpoint/restore support" 11645cb366bbSAdrian Reber select PROC_CHILDREN 11655cb366bbSAdrian Reber default n 11665cb366bbSAdrian Reber help 11675cb366bbSAdrian Reber Enables additional kernel features in a sake of checkpoint/restore. 11685cb366bbSAdrian Reber In particular it adds auxiliary prctl codes to setup process text, 11695cb366bbSAdrian Reber data and heap segment sizes, and a few additional /proc filesystem 11705cb366bbSAdrian Reber entries. 11715cb366bbSAdrian Reber 11725cb366bbSAdrian Reber If unsure, say N here. 11735cb366bbSAdrian Reber 11745091faa4SMike Galbraithconfig SCHED_AUTOGROUP 11755091faa4SMike Galbraith bool "Automatic process group scheduling" 11765091faa4SMike Galbraith select CGROUPS 11775091faa4SMike Galbraith select CGROUP_SCHED 11785091faa4SMike Galbraith select FAIR_GROUP_SCHED 11795091faa4SMike Galbraith help 11805091faa4SMike Galbraith This option optimizes the scheduler for common desktop workloads by 11815091faa4SMike Galbraith automatically creating and populating task groups. This separation 11825091faa4SMike Galbraith of workloads isolates aggressive CPU burners (like build jobs) from 11835091faa4SMike Galbraith desktop applications. Task group autogeneration is currently based 11845091faa4SMike Galbraith upon task session. 11855091faa4SMike Galbraith 11867af37becSDaniel Lezcanoconfig SYSFS_DEPRECATED 11875d6a4ea5SFerenc Wagner bool "Enable deprecated sysfs features to support old userspace tools" 11887af37becSDaniel Lezcano depends on SYSFS 11897af37becSDaniel Lezcano default n 11907af37becSDaniel Lezcano help 11917af37becSDaniel Lezcano This option adds code that switches the layout of the "block" class 11927af37becSDaniel Lezcano devices, to not show up in /sys/class/block/, but only in 11937af37becSDaniel Lezcano /sys/block/. 11947af37becSDaniel Lezcano 11957af37becSDaniel Lezcano This switch is only active when the sysfs.deprecated=1 boot option is 11967af37becSDaniel Lezcano passed or the SYSFS_DEPRECATED_V2 option is set. 11977af37becSDaniel Lezcano 11987af37becSDaniel Lezcano This option allows new kernels to run on old distributions and tools, 11997af37becSDaniel Lezcano which might get confused by /sys/class/block/. Since 2007/2008 all 12007af37becSDaniel Lezcano major distributions and tools handle this just fine. 12017af37becSDaniel Lezcano 12027af37becSDaniel Lezcano Recent distributions and userspace tools after 2009/2010 depend on 12037af37becSDaniel Lezcano the existence of /sys/class/block/, and will not work with this 12047af37becSDaniel Lezcano option enabled. 12057af37becSDaniel Lezcano 12067af37becSDaniel Lezcano Only if you are using a new kernel on an old distribution, you might 12077af37becSDaniel Lezcano need to say Y here. 12087af37becSDaniel Lezcano 12097af37becSDaniel Lezcanoconfig SYSFS_DEPRECATED_V2 12105d6a4ea5SFerenc Wagner bool "Enable deprecated sysfs features by default" 12117af37becSDaniel Lezcano default n 12127af37becSDaniel Lezcano depends on SYSFS 12137af37becSDaniel Lezcano depends on SYSFS_DEPRECATED 12147af37becSDaniel Lezcano help 12157af37becSDaniel Lezcano Enable deprecated sysfs by default. 12167af37becSDaniel Lezcano 12177af37becSDaniel Lezcano See the CONFIG_SYSFS_DEPRECATED option for more details about this 12187af37becSDaniel Lezcano option. 12197af37becSDaniel Lezcano 12207af37becSDaniel Lezcano Only if you are using a new kernel on an old distribution, you might 12217af37becSDaniel Lezcano need to say Y here. Even then, odds are you would not need it 12227af37becSDaniel Lezcano enabled, you can always pass the boot option if absolutely necessary. 12237af37becSDaniel Lezcano 12247af37becSDaniel Lezcanoconfig RELAY 12257af37becSDaniel Lezcano bool "Kernel->user space relay support (formerly relayfs)" 122626b5679eSPeter Zijlstra select IRQ_WORK 12277af37becSDaniel Lezcano help 12287af37becSDaniel Lezcano This option enables support for relay interface support in 12297af37becSDaniel Lezcano certain file systems (such as debugfs). 12307af37becSDaniel Lezcano It is designed to provide an efficient mechanism for tools and 12317af37becSDaniel Lezcano facilities to relay large amounts of data from kernel space to 12327af37becSDaniel Lezcano user space. 12337af37becSDaniel Lezcano 12347af37becSDaniel Lezcano If unsure, say N. 12357af37becSDaniel Lezcano 1236f991633dSDimitri Gorokhovikconfig BLK_DEV_INITRD 1237f991633dSDimitri Gorokhovik bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1238f991633dSDimitri Gorokhovik help 1239f991633dSDimitri Gorokhovik The initial RAM filesystem is a ramfs which is loaded by the 1240f991633dSDimitri Gorokhovik boot loader (loadlin or lilo) and that is mounted as root 1241f991633dSDimitri Gorokhovik before the normal boot procedure. It is typically used to 1242f991633dSDimitri Gorokhovik load modules needed to mount the "real" root file system, 12438c27ceffSMauro Carvalho Chehab etc. See <file:Documentation/admin-guide/initrd.rst> for details. 1244f991633dSDimitri Gorokhovik 1245f991633dSDimitri Gorokhovik If RAM disk support (BLK_DEV_RAM) is also included, this 1246f991633dSDimitri Gorokhovik also enables initial RAM disk (initrd) support and adds 1247f991633dSDimitri Gorokhovik 15 Kbytes (more on some other architectures) to the kernel size. 1248f991633dSDimitri Gorokhovik 1249f991633dSDimitri Gorokhovik If unsure say Y. 1250f991633dSDimitri Gorokhovik 1251c33df4eaSJean-Paul Samanif BLK_DEV_INITRD 1252c33df4eaSJean-Paul Saman 1253dbec4866SSam Ravnborgsource "usr/Kconfig" 1254dbec4866SSam Ravnborg 1255c33df4eaSJean-Paul Samanendif 1256c33df4eaSJean-Paul Saman 125776db5a27SMasami Hiramatsuconfig BOOT_CONFIG 125876db5a27SMasami Hiramatsu bool "Boot config support" 12592910b5aaSMasami Hiramatsu select BLK_DEV_INITRD 126076db5a27SMasami Hiramatsu help 126176db5a27SMasami Hiramatsu Extra boot config allows system admin to pass a config file as 126276db5a27SMasami Hiramatsu complemental extension of kernel cmdline when booting. 12630947db01SMasami Hiramatsu The boot config file must be attached at the end of initramfs 126485c46b78SMasami Hiramatsu with checksum, size and magic word. 12650947db01SMasami Hiramatsu See <file:Documentation/admin-guide/bootconfig.rst> for details. 126676db5a27SMasami Hiramatsu 126776db5a27SMasami Hiramatsu If unsure, say Y. 126876db5a27SMasami Hiramatsu 1269877417e6SArnd Bergmannchoice 1270877417e6SArnd Bergmann prompt "Compiler optimization level" 12712cc3ce24SUlf Magnusson default CC_OPTIMIZE_FOR_PERFORMANCE 1272877417e6SArnd Bergmann 1273877417e6SArnd Bergmannconfig CC_OPTIMIZE_FOR_PERFORMANCE 127415f5db60SMasahiro Yamada bool "Optimize for performance (-O2)" 1275877417e6SArnd Bergmann help 1276877417e6SArnd Bergmann This is the default optimization level for the kernel, building 1277877417e6SArnd Bergmann with the "-O2" compiler flag for best performance and most 1278877417e6SArnd Bergmann helpful compile-time warnings. 1279877417e6SArnd Bergmann 128015f5db60SMasahiro Yamadaconfig CC_OPTIMIZE_FOR_PERFORMANCE_O3 128115f5db60SMasahiro Yamada bool "Optimize more for performance (-O3)" 128215f5db60SMasahiro Yamada depends on ARC 1283c45b4f1fSLinus Torvalds help 128415f5db60SMasahiro Yamada Choosing this option will pass "-O3" to your compiler to optimize 128515f5db60SMasahiro Yamada the kernel yet more for performance. 1286c45b4f1fSLinus Torvalds 12875d20ee31SNicholas Pigginconfig CC_OPTIMIZE_FOR_SIZE 128815f5db60SMasahiro Yamada bool "Optimize for size (-Os)" 1289c45b4f1fSLinus Torvalds help 1290ce3b487fSMasahiro Yamada Choosing this option will pass "-Os" to your compiler resulting 1291ce3b487fSMasahiro Yamada in a smaller kernel. 1292c45b4f1fSLinus Torvalds 1293877417e6SArnd Bergmannendchoice 1294877417e6SArnd Bergmann 12955d20ee31SNicholas Pigginconfig HAVE_LD_DEAD_CODE_DATA_ELIMINATION 12965d20ee31SNicholas Piggin bool 12975d20ee31SNicholas Piggin help 12985d20ee31SNicholas Piggin This requires that the arch annotates or otherwise protects 12995d20ee31SNicholas Piggin its external entry points from being discarded. Linker scripts 13005d20ee31SNicholas Piggin must also merge .text.*, .data.*, and .bss.* correctly into 13015d20ee31SNicholas Piggin output sections. Care must be taken not to pull in unrelated 13025d20ee31SNicholas Piggin sections (e.g., '.text.init'). Typically '.' in section names 13035d20ee31SNicholas Piggin is used to distinguish them from label names / C identifiers. 13045d20ee31SNicholas Piggin 13055d20ee31SNicholas Pigginconfig LD_DEAD_CODE_DATA_ELIMINATION 13065d20ee31SNicholas Piggin bool "Dead code and data elimination (EXPERIMENTAL)" 13075d20ee31SNicholas Piggin depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION 13085d20ee31SNicholas Piggin depends on EXPERT 1309e85d1d65SMasahiro Yamada depends on $(cc-option,-ffunction-sections -fdata-sections) 1310e85d1d65SMasahiro Yamada depends on $(ld-option,--gc-sections) 13115d20ee31SNicholas Piggin help 13128b9d2712SMasahiro Yamada Enable this if you want to do dead code and data elimination with 13138b9d2712SMasahiro Yamada the linker by compiling with -ffunction-sections -fdata-sections, 13148b9d2712SMasahiro Yamada and linking with --gc-sections. 13155d20ee31SNicholas Piggin 13165d20ee31SNicholas Piggin This can reduce on disk and in-memory size of the kernel 13175d20ee31SNicholas Piggin code and static data, particularly for small configs and 13185d20ee31SNicholas Piggin on small systems. This has the possibility of introducing 13195d20ee31SNicholas Piggin silently broken kernel if the required annotations are not 13205d20ee31SNicholas Piggin present. This option is not well tested yet, so use at your 13215d20ee31SNicholas Piggin own risk. 13225d20ee31SNicholas Piggin 13230847062aSRandy Dunlapconfig SYSCTL 13240847062aSRandy Dunlap bool 13250847062aSRandy Dunlap 1326657a5209SMike Frysingerconfig HAVE_UID16 1327657a5209SMike Frysinger bool 1328657a5209SMike Frysinger 1329657a5209SMike Frysingerconfig SYSCTL_EXCEPTION_TRACE 1330657a5209SMike Frysinger bool 1331657a5209SMike Frysinger help 1332657a5209SMike Frysinger Enable support for /proc/sys/debug/exception-trace. 1333657a5209SMike Frysinger 1334657a5209SMike Frysingerconfig SYSCTL_ARCH_UNALIGN_NO_WARN 1335657a5209SMike Frysinger bool 1336657a5209SMike Frysinger help 1337657a5209SMike Frysinger Enable support for /proc/sys/kernel/ignore-unaligned-usertrap 1338657a5209SMike Frysinger Allows arch to define/use @no_unaligned_warning to possibly warn 1339657a5209SMike Frysinger about unaligned access emulation going on under the hood. 1340657a5209SMike Frysinger 1341657a5209SMike Frysingerconfig SYSCTL_ARCH_UNALIGN_ALLOW 1342657a5209SMike Frysinger bool 1343657a5209SMike Frysinger help 1344657a5209SMike Frysinger Enable support for /proc/sys/kernel/unaligned-trap 1345657a5209SMike Frysinger Allows arches to define/use @unaligned_enabled to runtime toggle 1346657a5209SMike Frysinger the unaligned access emulation. 1347657a5209SMike Frysinger see arch/parisc/kernel/unaligned.c for reference 1348657a5209SMike Frysinger 1349657a5209SMike Frysingerconfig HAVE_PCSPKR_PLATFORM 1350657a5209SMike Frysinger bool 1351657a5209SMike Frysinger 1352f89b7755SAlexei Starovoitov# interpreter that classic socket filters depend on 1353f89b7755SAlexei Starovoitovconfig BPF 1354f89b7755SAlexei Starovoitov bool 1355f89b7755SAlexei Starovoitov 13566a108a14SDavid Rientjesmenuconfig EXPERT 13576a108a14SDavid Rientjes bool "Configure standard kernel features (expert users)" 1358f505c553SJosh Triplett # Unhide debug options, to make the on-by-default options visible 1359f505c553SJosh Triplett select DEBUG_KERNEL 13601da177e4SLinus Torvalds help 13611da177e4SLinus Torvalds This option allows certain base kernel options and settings 13621da177e4SLinus Torvalds to be disabled or tweaked. This is for specialized 13631da177e4SLinus Torvalds environments which can tolerate a "non-standard" kernel. 13641da177e4SLinus Torvalds Only use this if you really know what you are doing. 13651da177e4SLinus Torvalds 1366ae81f9e3SChuck Ebbertconfig UID16 13676a108a14SDavid Rientjes bool "Enable 16-bit UID system calls" if EXPERT 13682813893fSIulia Manda depends on HAVE_UID16 && MULTIUSER 1369ae81f9e3SChuck Ebbert default y 1370ae81f9e3SChuck Ebbert help 1371ae81f9e3SChuck Ebbert This enables the legacy 16-bit UID syscall wrappers. 1372ae81f9e3SChuck Ebbert 13732813893fSIulia Mandaconfig MULTIUSER 13742813893fSIulia Manda bool "Multiple users, groups and capabilities support" if EXPERT 13752813893fSIulia Manda default y 13762813893fSIulia Manda help 13772813893fSIulia Manda This option enables support for non-root users, groups and 13782813893fSIulia Manda capabilities. 13792813893fSIulia Manda 13802813893fSIulia Manda If you say N here, all processes will run with UID 0, GID 0, and all 13812813893fSIulia Manda possible capabilities. Saying N here also compiles out support for 13822813893fSIulia Manda system calls related to UIDs, GIDs, and capabilities, such as setuid, 13832813893fSIulia Manda setgid, and capset. 13842813893fSIulia Manda 13852813893fSIulia Manda If unsure, say Y here. 13862813893fSIulia Manda 1387f6187769SFabian Frederickconfig SGETMASK_SYSCALL 1388f6187769SFabian Frederick bool "sgetmask/ssetmask syscalls support" if EXPERT 1389a687a533SArnd Bergmann def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH 1390a7f7f624SMasahiro Yamada help 1391f6187769SFabian Frederick sys_sgetmask and sys_ssetmask are obsolete system calls 1392f6187769SFabian Frederick no longer supported in libc but still enabled by default in some 1393f6187769SFabian Frederick architectures. 1394f6187769SFabian Frederick 1395f6187769SFabian Frederick If unsure, leave the default option here. 1396f6187769SFabian Frederick 13976af9f7bfSFabian Frederickconfig SYSFS_SYSCALL 13986af9f7bfSFabian Frederick bool "Sysfs syscall support" if EXPERT 13996af9f7bfSFabian Frederick default y 1400a7f7f624SMasahiro Yamada help 14016af9f7bfSFabian Frederick sys_sysfs is an obsolete system call no longer supported in libc. 14026af9f7bfSFabian Frederick Note that disabling this option is more secure but might break 14036af9f7bfSFabian Frederick compatibility with some systems. 14046af9f7bfSFabian Frederick 14056af9f7bfSFabian Frederick If unsure say Y here. 14066af9f7bfSFabian Frederick 1407d1b069f5SRandy Dunlapconfig FHANDLE 1408d1b069f5SRandy Dunlap bool "open by fhandle syscalls" if EXPERT 1409d1b069f5SRandy Dunlap select EXPORTFS 1410d1b069f5SRandy Dunlap default y 1411d1b069f5SRandy Dunlap help 1412d1b069f5SRandy Dunlap If you say Y here, a user level program will be able to map 1413d1b069f5SRandy Dunlap file names to handle and then later use the handle for 1414d1b069f5SRandy Dunlap different file system operations. This is useful in implementing 1415d1b069f5SRandy Dunlap userspace file servers, which now track files using handles instead 1416d1b069f5SRandy Dunlap of names. The handle would remain the same even if file names 1417d1b069f5SRandy Dunlap get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 1418d1b069f5SRandy Dunlap syscalls. 1419d1b069f5SRandy Dunlap 1420baa73d9eSNicolas Pitreconfig POSIX_TIMERS 1421baa73d9eSNicolas Pitre bool "Posix Clocks & timers" if EXPERT 1422baa73d9eSNicolas Pitre default y 1423baa73d9eSNicolas Pitre help 1424baa73d9eSNicolas Pitre This includes native support for POSIX timers to the kernel. 1425baa73d9eSNicolas Pitre Some embedded systems have no use for them and therefore they 1426baa73d9eSNicolas Pitre can be configured out to reduce the size of the kernel image. 1427baa73d9eSNicolas Pitre 1428baa73d9eSNicolas Pitre When this option is disabled, the following syscalls won't be 1429baa73d9eSNicolas Pitre available: timer_create, timer_gettime: timer_getoverrun, 1430baa73d9eSNicolas Pitre timer_settime, timer_delete, clock_adjtime, getitimer, 1431baa73d9eSNicolas Pitre setitimer, alarm. Furthermore, the clock_settime, clock_gettime, 1432baa73d9eSNicolas Pitre clock_getres and clock_nanosleep syscalls will be limited to 1433baa73d9eSNicolas Pitre CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only. 1434baa73d9eSNicolas Pitre 1435baa73d9eSNicolas Pitre If unsure say y. 1436baa73d9eSNicolas Pitre 1437d59745ceSMatt Mackallconfig PRINTK 1438d59745ceSMatt Mackall default y 14396a108a14SDavid Rientjes bool "Enable support for printk" if EXPERT 144074876a98SFrederic Weisbecker select IRQ_WORK 1441d59745ceSMatt Mackall help 1442d59745ceSMatt Mackall This option enables normal printk support. Removing it 1443d59745ceSMatt Mackall eliminates most of the message strings from the kernel image 1444d59745ceSMatt Mackall and makes the kernel more or less silent. As this makes it 1445d59745ceSMatt Mackall very difficult to diagnose system problems, saying N here is 1446d59745ceSMatt Mackall strongly discouraged. 1447d59745ceSMatt Mackall 144842a0bb3fSPetr Mladekconfig PRINTK_NMI 144942a0bb3fSPetr Mladek def_bool y 145042a0bb3fSPetr Mladek depends on PRINTK 145142a0bb3fSPetr Mladek depends on HAVE_NMI 145242a0bb3fSPetr Mladek 1453c8538a7aSMatt Mackallconfig BUG 14546a108a14SDavid Rientjes bool "BUG() support" if EXPERT 1455c8538a7aSMatt Mackall default y 1456c8538a7aSMatt Mackall help 1457c8538a7aSMatt Mackall Disabling this option eliminates support for BUG and WARN, reducing 1458c8538a7aSMatt Mackall the size of your kernel image and potentially quietly ignoring 1459c8538a7aSMatt Mackall numerous fatal conditions. You should only consider disabling this 1460c8538a7aSMatt Mackall option for embedded systems with no facilities for reporting errors. 1461c8538a7aSMatt Mackall Just say Y. 1462c8538a7aSMatt Mackall 1463708e9a79SMatt Mackallconfig ELF_CORE 1464046d662fSAlex Kelly depends on COREDUMP 1465708e9a79SMatt Mackall default y 14666a108a14SDavid Rientjes bool "Enable ELF core dumps" if EXPERT 1467708e9a79SMatt Mackall help 1468708e9a79SMatt Mackall Enable support for generating core dumps. Disabling saves about 4k. 1469708e9a79SMatt Mackall 14708761f1abSRalf Baechle 1471e5e1d3cbSStas Sergeevconfig PCSPKR_PLATFORM 14726a108a14SDavid Rientjes bool "Enable PC-Speaker support" if EXPERT 14738761f1abSRalf Baechle depends on HAVE_PCSPKR_PLATFORM 147415f304b6SRalf Baechle select I8253_LOCK 1475e5e1d3cbSStas Sergeev default y 1476e5e1d3cbSStas Sergeev help 1477e5e1d3cbSStas Sergeev This option allows to disable the internal PC-Speaker 1478e5e1d3cbSStas Sergeev support, saving some memory. 1479e5e1d3cbSStas Sergeev 14801da177e4SLinus Torvaldsconfig BASE_FULL 14811da177e4SLinus Torvalds default y 14826a108a14SDavid Rientjes bool "Enable full-sized data structures for core" if EXPERT 14831da177e4SLinus Torvalds help 14841da177e4SLinus Torvalds Disabling this option reduces the size of miscellaneous core 14851da177e4SLinus Torvalds kernel data structures. This saves memory on small machines, 14861da177e4SLinus Torvalds but may reduce performance. 14871da177e4SLinus Torvalds 14881da177e4SLinus Torvaldsconfig FUTEX 14896a108a14SDavid Rientjes bool "Enable futex support" if EXPERT 14901da177e4SLinus Torvalds default y 1491bc2eecd7SNicolas Pitre imply RT_MUTEXES 14921da177e4SLinus Torvalds help 14931da177e4SLinus Torvalds Disabling this option will cause the kernel to be built without 14941da177e4SLinus Torvalds support for "fast userspace mutexes". The resulting kernel may not 14951da177e4SLinus Torvalds run glibc-based applications correctly. 14961da177e4SLinus Torvalds 1497bc2eecd7SNicolas Pitreconfig FUTEX_PI 1498bc2eecd7SNicolas Pitre bool 1499bc2eecd7SNicolas Pitre depends on FUTEX && RT_MUTEXES 1500bc2eecd7SNicolas Pitre default y 1501bc2eecd7SNicolas Pitre 150203b8c7b6SHeiko Carstensconfig HAVE_FUTEX_CMPXCHG 150303b8c7b6SHeiko Carstens bool 150462b4d204SJosh Triplett depends on FUTEX 150503b8c7b6SHeiko Carstens help 150603b8c7b6SHeiko Carstens Architectures should select this if futex_atomic_cmpxchg_inatomic() 150703b8c7b6SHeiko Carstens is implemented and always working. This removes a couple of runtime 150803b8c7b6SHeiko Carstens checks. 150903b8c7b6SHeiko Carstens 15101da177e4SLinus Torvaldsconfig EPOLL 15116a108a14SDavid Rientjes bool "Enable eventpoll support" if EXPERT 15121da177e4SLinus Torvalds default y 15131da177e4SLinus Torvalds help 15141da177e4SLinus Torvalds Disabling this option will cause the kernel to be built without 15151da177e4SLinus Torvalds support for epoll family of system calls. 15161da177e4SLinus Torvalds 1517fba2afaaSDavide Libenziconfig SIGNALFD 15186a108a14SDavid Rientjes bool "Enable signalfd() system call" if EXPERT 1519fba2afaaSDavide Libenzi default y 1520fba2afaaSDavide Libenzi help 1521fba2afaaSDavide Libenzi Enable the signalfd() system call that allows to receive signals 1522fba2afaaSDavide Libenzi on a file descriptor. 1523fba2afaaSDavide Libenzi 1524fba2afaaSDavide Libenzi If unsure, say Y. 1525fba2afaaSDavide Libenzi 1526b215e283SDavide Libenziconfig TIMERFD 15276a108a14SDavid Rientjes bool "Enable timerfd() system call" if EXPERT 1528b215e283SDavide Libenzi default y 1529b215e283SDavide Libenzi help 1530b215e283SDavide Libenzi Enable the timerfd() system call that allows to receive timer 1531b215e283SDavide Libenzi events on a file descriptor. 1532b215e283SDavide Libenzi 1533b215e283SDavide Libenzi If unsure, say Y. 1534b215e283SDavide Libenzi 1535e1ad7468SDavide Libenziconfig EVENTFD 15366a108a14SDavid Rientjes bool "Enable eventfd() system call" if EXPERT 1537e1ad7468SDavide Libenzi default y 1538e1ad7468SDavide Libenzi help 1539e1ad7468SDavide Libenzi Enable the eventfd() system call that allows to receive both 1540e1ad7468SDavide Libenzi kernel notification (ie. KAIO) or userspace notifications. 1541e1ad7468SDavide Libenzi 1542e1ad7468SDavide Libenzi If unsure, say Y. 1543e1ad7468SDavide Libenzi 15441da177e4SLinus Torvaldsconfig SHMEM 15456a108a14SDavid Rientjes bool "Use full shmem filesystem" if EXPERT 15461da177e4SLinus Torvalds default y 15471da177e4SLinus Torvalds depends on MMU 15481da177e4SLinus Torvalds help 15491da177e4SLinus Torvalds The shmem is an internal filesystem used to manage shared memory. 15501da177e4SLinus Torvalds It is backed by swap and manages resource limits. It is also exported 15511da177e4SLinus Torvalds to userspace as tmpfs if TMPFS is enabled. Disabling this 15521da177e4SLinus Torvalds option replaces shmem and tmpfs with the much simpler ramfs code, 15531da177e4SLinus Torvalds which may be appropriate on small systems without swap. 15541da177e4SLinus Torvalds 1555ebf3f09cSThomas Petazzoniconfig AIO 15566a108a14SDavid Rientjes bool "Enable AIO support" if EXPERT 1557ebf3f09cSThomas Petazzoni default y 1558ebf3f09cSThomas Petazzoni help 1559ebf3f09cSThomas Petazzoni This option enables POSIX asynchronous I/O which may by used 1560ebf3f09cSThomas Petazzoni by some high performance threaded applications. Disabling 1561ebf3f09cSThomas Petazzoni this option saves about 7k. 1562ebf3f09cSThomas Petazzoni 15632b188cc1SJens Axboeconfig IO_URING 15642b188cc1SJens Axboe bool "Enable IO uring support" if EXPERT 1565561fb04aSJens Axboe select IO_WQ 15662b188cc1SJens Axboe default y 15672b188cc1SJens Axboe help 15682b188cc1SJens Axboe This option enables support for the io_uring interface, enabling 15692b188cc1SJens Axboe applications to submit and complete IO through submission and 15702b188cc1SJens Axboe completion rings that are shared between the kernel and application. 15712b188cc1SJens Axboe 1572d3ac21caSJosh Triplettconfig ADVISE_SYSCALLS 1573d3ac21caSJosh Triplett bool "Enable madvise/fadvise syscalls" if EXPERT 1574d3ac21caSJosh Triplett default y 1575d3ac21caSJosh Triplett help 1576d3ac21caSJosh Triplett This option enables the madvise and fadvise syscalls, used by 1577d3ac21caSJosh Triplett applications to advise the kernel about their future memory or file 1578d3ac21caSJosh Triplett usage, improving performance. If building an embedded system where no 1579d3ac21caSJosh Triplett applications use these syscalls, you can disable this option to save 1580d3ac21caSJosh Triplett space. 1581d3ac21caSJosh Triplett 15825a281062SAndrea Arcangeliconfig HAVE_ARCH_USERFAULTFD_WP 15835a281062SAndrea Arcangeli bool 15845a281062SAndrea Arcangeli help 15855a281062SAndrea Arcangeli Arch has userfaultfd write protection support 15865a281062SAndrea Arcangeli 15875b25b13aSMathieu Desnoyersconfig MEMBARRIER 15885b25b13aSMathieu Desnoyers bool "Enable membarrier() system call" if EXPERT 15895b25b13aSMathieu Desnoyers default y 15905b25b13aSMathieu Desnoyers help 15915b25b13aSMathieu Desnoyers Enable the membarrier() system call that allows issuing memory 15925b25b13aSMathieu Desnoyers barriers across all running threads, which can be used to distribute 15935b25b13aSMathieu Desnoyers the cost of user-space memory barriers asymmetrically by transforming 15945b25b13aSMathieu Desnoyers pairs of memory barriers into pairs consisting of membarrier() and a 15955b25b13aSMathieu Desnoyers compiler barrier. 15965b25b13aSMathieu Desnoyers 15975b25b13aSMathieu Desnoyers If unsure, say Y. 15985b25b13aSMathieu Desnoyers 1599d1b069f5SRandy Dunlapconfig KALLSYMS 1600d1b069f5SRandy Dunlap bool "Load all symbols for debugging/ksymoops" if EXPERT 1601d1b069f5SRandy Dunlap default y 1602d1b069f5SRandy Dunlap help 1603d1b069f5SRandy Dunlap Say Y here to let the kernel print out symbolic crash information and 1604d1b069f5SRandy Dunlap symbolic stack backtraces. This increases the size of the kernel 1605d1b069f5SRandy Dunlap somewhat, as all symbols have to be loaded into the kernel image. 1606d1b069f5SRandy Dunlap 1607d1b069f5SRandy Dunlapconfig KALLSYMS_ALL 1608d1b069f5SRandy Dunlap bool "Include all symbols in kallsyms" 1609d1b069f5SRandy Dunlap depends on DEBUG_KERNEL && KALLSYMS 1610d1b069f5SRandy Dunlap help 1611d1b069f5SRandy Dunlap Normally kallsyms only contains the symbols of functions for nicer 1612d1b069f5SRandy Dunlap OOPS messages and backtraces (i.e., symbols from the text and inittext 1613d1b069f5SRandy Dunlap sections). This is sufficient for most cases. And only in very rare 1614d1b069f5SRandy Dunlap cases (e.g., when a debugger is used) all symbols are required (e.g., 1615d1b069f5SRandy Dunlap names of variables from the data sections, etc). 1616d1b069f5SRandy Dunlap 1617d1b069f5SRandy Dunlap This option makes sure that all symbols are loaded into the kernel 1618d1b069f5SRandy Dunlap image (i.e., symbols from all sections) in cost of increased kernel 1619d1b069f5SRandy Dunlap size (depending on the kernel configuration, it may be 300KiB or 1620d1b069f5SRandy Dunlap something like this). 1621d1b069f5SRandy Dunlap 1622d1b069f5SRandy Dunlap Say N unless you really need all symbols. 1623d1b069f5SRandy Dunlap 1624d1b069f5SRandy Dunlapconfig KALLSYMS_ABSOLUTE_PERCPU 1625d1b069f5SRandy Dunlap bool 1626d1b069f5SRandy Dunlap depends on KALLSYMS 1627d1b069f5SRandy Dunlap default X86_64 && SMP 1628d1b069f5SRandy Dunlap 1629d1b069f5SRandy Dunlapconfig KALLSYMS_BASE_RELATIVE 1630d1b069f5SRandy Dunlap bool 1631d1b069f5SRandy Dunlap depends on KALLSYMS 1632a687a533SArnd Bergmann default !IA64 1633d1b069f5SRandy Dunlap help 1634d1b069f5SRandy Dunlap Instead of emitting them as absolute values in the native word size, 1635d1b069f5SRandy Dunlap emit the symbol references in the kallsyms table as 32-bit entries, 1636d1b069f5SRandy Dunlap each containing a relative value in the range [base, base + U32_MAX] 1637d1b069f5SRandy Dunlap or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either 1638d1b069f5SRandy Dunlap an absolute value in the range [0, S32_MAX] or a relative value in the 1639d1b069f5SRandy Dunlap range [base, base + S32_MAX], where base is the lowest relative symbol 1640d1b069f5SRandy Dunlap address encountered in the image. 1641d1b069f5SRandy Dunlap 1642d1b069f5SRandy Dunlap On 64-bit builds, this reduces the size of the address table by 50%, 1643d1b069f5SRandy Dunlap but more importantly, it results in entries whose values are build 1644d1b069f5SRandy Dunlap time constants, and no relocation pass is required at runtime to fix 1645d1b069f5SRandy Dunlap up the entries based on the runtime load address of the kernel. 1646d1b069f5SRandy Dunlap 1647d1b069f5SRandy Dunlap# end of the "standard kernel features (expert users)" menu 1648d1b069f5SRandy Dunlap 1649d1b069f5SRandy Dunlap# syscall, maps, verifier 1650fc611f47SKP Singh 1651fc611f47SKP Singhconfig BPF_LSM 1652fc611f47SKP Singh bool "LSM Instrumentation with BPF" 16534edf16b7SKP Singh depends on BPF_EVENTS 1654fc611f47SKP Singh depends on BPF_SYSCALL 1655fc611f47SKP Singh depends on SECURITY 1656fc611f47SKP Singh depends on BPF_JIT 1657fc611f47SKP Singh help 1658fc611f47SKP Singh Enables instrumentation of the security hooks with eBPF programs for 1659fc611f47SKP Singh implementing dynamic MAC and Audit Policies. 1660fc611f47SKP Singh 1661fc611f47SKP Singh If you are unsure how to answer this question, answer N. 1662fc611f47SKP Singh 1663d1b069f5SRandy Dunlapconfig BPF_SYSCALL 1664d1b069f5SRandy Dunlap bool "Enable bpf() system call" 1665d1b069f5SRandy Dunlap select BPF 1666bae77c5eSSong Liu select IRQ_WORK 1667d1b069f5SRandy Dunlap default n 1668d1b069f5SRandy Dunlap help 1669d1b069f5SRandy Dunlap Enable the bpf() system call that allows to manipulate eBPF 1670d1b069f5SRandy Dunlap programs and maps via file descriptors. 1671d1b069f5SRandy Dunlap 167281c22041SDaniel Borkmannconfig ARCH_WANT_DEFAULT_BPF_JIT 167381c22041SDaniel Borkmann bool 167481c22041SDaniel Borkmann 1675290af866SAlexei Starovoitovconfig BPF_JIT_ALWAYS_ON 1676290af866SAlexei Starovoitov bool "Permanently enable BPF JIT and remove BPF interpreter" 1677290af866SAlexei Starovoitov depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT 1678290af866SAlexei Starovoitov help 1679290af866SAlexei Starovoitov Enables BPF JIT and removes BPF interpreter to avoid 1680290af866SAlexei Starovoitov speculative execution of BPF instructions by the interpreter 1681290af866SAlexei Starovoitov 168281c22041SDaniel Borkmannconfig BPF_JIT_DEFAULT_ON 168381c22041SDaniel Borkmann def_bool ARCH_WANT_DEFAULT_BPF_JIT || BPF_JIT_ALWAYS_ON 168481c22041SDaniel Borkmann depends on HAVE_EBPF_JIT && BPF_JIT 168581c22041SDaniel Borkmann 1686d1b069f5SRandy Dunlapconfig USERFAULTFD 1687d1b069f5SRandy Dunlap bool "Enable userfaultfd() system call" 1688d1b069f5SRandy Dunlap depends on MMU 1689d1b069f5SRandy Dunlap help 1690d1b069f5SRandy Dunlap Enable the userfaultfd() system call that allows to intercept and 1691d1b069f5SRandy Dunlap handle page faults in userland. 1692d1b069f5SRandy Dunlap 16933ccfebedSMathieu Desnoyersconfig ARCH_HAS_MEMBARRIER_CALLBACKS 16943ccfebedSMathieu Desnoyers bool 16953ccfebedSMathieu Desnoyers 169670216e18SMathieu Desnoyersconfig ARCH_HAS_MEMBARRIER_SYNC_CORE 169770216e18SMathieu Desnoyers bool 169870216e18SMathieu Desnoyers 1699d7822b1eSMathieu Desnoyersconfig RSEQ 1700d7822b1eSMathieu Desnoyers bool "Enable rseq() system call" if EXPERT 1701d7822b1eSMathieu Desnoyers default y 1702d7822b1eSMathieu Desnoyers depends on HAVE_RSEQ 1703d7822b1eSMathieu Desnoyers select MEMBARRIER 1704d7822b1eSMathieu Desnoyers help 1705d7822b1eSMathieu Desnoyers Enable the restartable sequences system call. It provides a 1706d7822b1eSMathieu Desnoyers user-space cache for the current CPU number value, which 1707d7822b1eSMathieu Desnoyers speeds up getting the current CPU number from user-space, 1708d7822b1eSMathieu Desnoyers as well as an ABI to speed up user-space operations on 1709d7822b1eSMathieu Desnoyers per-CPU data. 1710d7822b1eSMathieu Desnoyers 1711d7822b1eSMathieu Desnoyers If unsure, say Y. 1712d7822b1eSMathieu Desnoyers 1713d7822b1eSMathieu Desnoyersconfig DEBUG_RSEQ 1714d7822b1eSMathieu Desnoyers default n 1715d7822b1eSMathieu Desnoyers bool "Enabled debugging of rseq() system call" if EXPERT 1716d7822b1eSMathieu Desnoyers depends on RSEQ && DEBUG_KERNEL 1717d7822b1eSMathieu Desnoyers help 1718d7822b1eSMathieu Desnoyers Enable extra debugging checks for the rseq system call. 1719d7822b1eSMathieu Desnoyers 1720d7822b1eSMathieu Desnoyers If unsure, say N. 1721d7822b1eSMathieu Desnoyers 17226befe5f6SRandy Dunlapconfig EMBEDDED 17236befe5f6SRandy Dunlap bool "Embedded system" 17245d2acfc7SJosh Triplett option allnoconfig_y 17256befe5f6SRandy Dunlap select EXPERT 17266befe5f6SRandy Dunlap help 17276befe5f6SRandy Dunlap This option should be enabled if compiling the kernel for 17286befe5f6SRandy Dunlap an embedded system so certain expert options are available 17296befe5f6SRandy Dunlap for configuration. 17306befe5f6SRandy Dunlap 1731cdd6c482SIngo Molnarconfig HAVE_PERF_EVENTS 17320793a61dSThomas Gleixner bool 1733018df72dSMike Frysinger help 1734018df72dSMike Frysinger See tools/perf/design.txt for details. 17350793a61dSThomas Gleixner 1736906010b2SPeter Zijlstraconfig PERF_USE_VMALLOC 1737906010b2SPeter Zijlstra bool 1738906010b2SPeter Zijlstra help 1739906010b2SPeter Zijlstra See tools/perf/design.txt for details 1740906010b2SPeter Zijlstra 1741ad90a3deSWilliam Breathitt Grayconfig PC104 1742424529fbSWilliam Breathitt Gray bool "PC/104 support" if EXPERT 1743ad90a3deSWilliam Breathitt Gray help 1744ad90a3deSWilliam Breathitt Gray Expose PC/104 form factor device drivers and options available for 1745ad90a3deSWilliam Breathitt Gray selection and configuration. Enable this option if your target 1746ad90a3deSWilliam Breathitt Gray machine has a PC/104 bus. 1747ad90a3deSWilliam Breathitt Gray 174857c0c15bSIngo Molnarmenu "Kernel Performance Events And Counters" 17490793a61dSThomas Gleixner 1750cdd6c482SIngo Molnarconfig PERF_EVENTS 175157c0c15bSIngo Molnar bool "Kernel performance events and counters" 1752392d65a9SRobert Richter default y if PROFILING 1753cdd6c482SIngo Molnar depends on HAVE_PERF_EVENTS 1754e360adbeSPeter Zijlstra select IRQ_WORK 175583fe27eaSPranith Kumar select SRCU 17560793a61dSThomas Gleixner help 175757c0c15bSIngo Molnar Enable kernel support for various performance events provided 175857c0c15bSIngo Molnar by software and hardware. 17590793a61dSThomas Gleixner 1760dd77038dSThadeu Lima de Souza Cascardo Software events are supported either built-in or via the 176157c0c15bSIngo Molnar use of generic tracepoints. 176257c0c15bSIngo Molnar 176357c0c15bSIngo Molnar Most modern CPUs support performance events via performance 176457c0c15bSIngo Molnar counter registers. These registers count the number of certain 17650793a61dSThomas Gleixner types of hw events: such as instructions executed, cachemisses 17660793a61dSThomas Gleixner suffered, or branches mis-predicted - without slowing down the 17670793a61dSThomas Gleixner kernel or applications. These registers can also trigger interrupts 17680793a61dSThomas Gleixner when a threshold number of events have passed - and can thus be 17690793a61dSThomas Gleixner used to profile the code that runs on that CPU. 17700793a61dSThomas Gleixner 177157c0c15bSIngo Molnar The Linux Performance Event subsystem provides an abstraction of 1772dd77038dSThadeu Lima de Souza Cascardo these software and hardware event capabilities, available via a 177357c0c15bSIngo Molnar system call and used by the "perf" utility in tools/perf/. It 17740793a61dSThomas Gleixner provides per task and per CPU counters, and it provides event 17750793a61dSThomas Gleixner capabilities on top of those. 17760793a61dSThomas Gleixner 17770793a61dSThomas Gleixner Say Y if unsure. 17780793a61dSThomas Gleixner 1779906010b2SPeter Zijlstraconfig DEBUG_PERF_USE_VMALLOC 1780906010b2SPeter Zijlstra default n 1781906010b2SPeter Zijlstra bool "Debug: use vmalloc to back perf mmap() buffers" 1782cb307113SMichael Ellerman depends on PERF_EVENTS && DEBUG_KERNEL && !PPC 1783906010b2SPeter Zijlstra select PERF_USE_VMALLOC 1784906010b2SPeter Zijlstra help 1785906010b2SPeter Zijlstra Use vmalloc memory to back perf mmap() buffers. 1786906010b2SPeter Zijlstra 1787906010b2SPeter Zijlstra Mostly useful for debugging the vmalloc code on platforms 1788906010b2SPeter Zijlstra that don't require it. 1789906010b2SPeter Zijlstra 1790906010b2SPeter Zijlstra Say N if unsure. 1791906010b2SPeter Zijlstra 17920793a61dSThomas Gleixnerendmenu 17930793a61dSThomas Gleixner 1794f8891e5eSChristoph Lameterconfig VM_EVENT_COUNTERS 1795f8891e5eSChristoph Lameter default y 17966a108a14SDavid Rientjes bool "Enable VM event counters for /proc/vmstat" if EXPERT 1797f8891e5eSChristoph Lameter help 17982aea4fb6SPaul Jackson VM event counters are needed for event counts to be shown. 17992aea4fb6SPaul Jackson This option allows the disabling of the VM event counters 18006a108a14SDavid Rientjes on EXPERT systems. /proc/vmstat will only show page counts 18012aea4fb6SPaul Jackson if VM event counters are disabled. 1802f8891e5eSChristoph Lameter 180341ecc55bSChristoph Lameterconfig SLUB_DEBUG 180441ecc55bSChristoph Lameter default y 18056a108a14SDavid Rientjes bool "Enable SLUB debugging support" if EXPERT 1806f6acb635SChristoph Lameter depends on SLUB && SYSFS 180741ecc55bSChristoph Lameter help 180841ecc55bSChristoph Lameter SLUB has extensive debug support features. Disabling these can 180941ecc55bSChristoph Lameter result in significant savings in code size. This also disables 181041ecc55bSChristoph Lameter SLUB sysfs support. /sys/slab will not exist and there will be 181141ecc55bSChristoph Lameter no support for cache validation etc. 181241ecc55bSChristoph Lameter 18131663f26dSTejun Heoconfig SLUB_MEMCG_SYSFS_ON 18141663f26dSTejun Heo default n 18151663f26dSTejun Heo bool "Enable memcg SLUB sysfs support by default" if EXPERT 18161663f26dSTejun Heo depends on SLUB && SYSFS && MEMCG 18171663f26dSTejun Heo help 18181663f26dSTejun Heo SLUB creates a directory under /sys/kernel/slab for each 18191663f26dSTejun Heo allocation cache to host info and debug files. If memory 18201663f26dSTejun Heo cgroup is enabled, each cache can have per memory cgroup 18211663f26dSTejun Heo caches. SLUB can create the same sysfs directories for these 18221663f26dSTejun Heo caches under /sys/kernel/slab/CACHE/cgroup but it can lead 18231663f26dSTejun Heo to a very high number of debug files being created. This is 18241663f26dSTejun Heo controlled by slub_memcg_sysfs boot parameter and this 18251663f26dSTejun Heo config option determines the parameter's default value. 18261663f26dSTejun Heo 1827b943c460SRandy Dunlapconfig COMPAT_BRK 1828b943c460SRandy Dunlap bool "Disable heap randomization" 1829b943c460SRandy Dunlap default y 1830b943c460SRandy Dunlap help 1831b943c460SRandy Dunlap Randomizing heap placement makes heap exploits harder, but it 1832b943c460SRandy Dunlap also breaks ancient binaries (including anything libc5 based). 1833b943c460SRandy Dunlap This option changes the bootup default to heap randomization 1834692105b8SMatt LaPlante disabled, and can be overridden at runtime by setting 1835b943c460SRandy Dunlap /proc/sys/kernel/randomize_va_space to 2. 1836b943c460SRandy Dunlap 1837b943c460SRandy Dunlap On non-ancient distros (post-2000 ones) N is usually a safe choice. 1838b943c460SRandy Dunlap 183981819f0fSChristoph Lameterchoice 184081819f0fSChristoph Lameter prompt "Choose SLAB allocator" 1841a0acd820SChristoph Lameter default SLUB 184281819f0fSChristoph Lameter help 184381819f0fSChristoph Lameter This option allows to select a slab allocator. 184481819f0fSChristoph Lameter 184581819f0fSChristoph Lameterconfig SLAB 184681819f0fSChristoph Lameter bool "SLAB" 184704385fc5SKees Cook select HAVE_HARDENED_USERCOPY_ALLOCATOR 184881819f0fSChristoph Lameter help 184981819f0fSChristoph Lameter The regular slab allocator that is established and known to work 185034013886SChristoph Lameter well in all environments. It organizes cache hot objects in 185102f56210SSimon Arlott per cpu and per node queues. 185281819f0fSChristoph Lameter 185381819f0fSChristoph Lameterconfig SLUB 185481819f0fSChristoph Lameter bool "SLUB (Unqueued Allocator)" 1855ed18adc1SKees Cook select HAVE_HARDENED_USERCOPY_ALLOCATOR 185681819f0fSChristoph Lameter help 185781819f0fSChristoph Lameter SLUB is a slab allocator that minimizes cache line usage 185881819f0fSChristoph Lameter instead of managing queues of cached objects (SLAB approach). 185981819f0fSChristoph Lameter Per cpu caching is realized using slabs of objects instead 186081819f0fSChristoph Lameter of queues of objects. SLUB can use memory efficiently 186102f56210SSimon Arlott and has enhanced diagnostics. SLUB is the default choice for 186202f56210SSimon Arlott a slab allocator. 186381819f0fSChristoph Lameter 186481819f0fSChristoph Lameterconfig SLOB 18656a108a14SDavid Rientjes depends on EXPERT 186681819f0fSChristoph Lameter bool "SLOB (Simple Allocator)" 186781819f0fSChristoph Lameter help 186837291458SMatt Mackall SLOB replaces the stock allocator with a drastically simpler 186937291458SMatt Mackall allocator. SLOB is generally more space efficient but 187037291458SMatt Mackall does not perform as well on large systems. 187181819f0fSChristoph Lameter 187281819f0fSChristoph Lameterendchoice 187381819f0fSChristoph Lameter 18747660a6fdSKees Cookconfig SLAB_MERGE_DEFAULT 18757660a6fdSKees Cook bool "Allow slab caches to be merged" 18767660a6fdSKees Cook default y 18777660a6fdSKees Cook help 18787660a6fdSKees Cook For reduced kernel memory fragmentation, slab caches can be 18797660a6fdSKees Cook merged when they share the same size and other characteristics. 18807660a6fdSKees Cook This carries a risk of kernel heap overflows being able to 18817660a6fdSKees Cook overwrite objects from merged caches (and more easily control 18827660a6fdSKees Cook cache layout), which makes such heap attacks easier to exploit 18837660a6fdSKees Cook by attackers. By keeping caches unmerged, these kinds of exploits 18847660a6fdSKees Cook can usually only damage objects in the same cache. To disable 18857660a6fdSKees Cook merging at runtime, "slab_nomerge" can be passed on the kernel 18867660a6fdSKees Cook command line. 18877660a6fdSKees Cook 1888c7ce4f60SThomas Garnierconfig SLAB_FREELIST_RANDOM 1889c7ce4f60SThomas Garnier default n 1890210e7a43SThomas Garnier depends on SLAB || SLUB 1891c7ce4f60SThomas Garnier bool "SLAB freelist randomization" 1892c7ce4f60SThomas Garnier help 1893210e7a43SThomas Garnier Randomizes the freelist order used on creating new pages. This 1894c7ce4f60SThomas Garnier security feature reduces the predictability of the kernel slab 1895c7ce4f60SThomas Garnier allocator against heap overflows. 1896c7ce4f60SThomas Garnier 18972482ddecSKees Cookconfig SLAB_FREELIST_HARDENED 18982482ddecSKees Cook bool "Harden slab freelist metadata" 18992482ddecSKees Cook depends on SLUB 19002482ddecSKees Cook help 19012482ddecSKees Cook Many kernel heap attacks try to target slab cache metadata and 19022482ddecSKees Cook other infrastructure. This options makes minor performance 190392bae787SKees Cook sacrifices to harden the kernel slab allocator against common 19042482ddecSKees Cook freelist exploit methods. 19052482ddecSKees Cook 1906e900a918SDan Williamsconfig SHUFFLE_PAGE_ALLOCATOR 1907e900a918SDan Williams bool "Page allocator randomization" 1908e900a918SDan Williams default SLAB_FREELIST_RANDOM && ACPI_NUMA 1909e900a918SDan Williams help 1910e900a918SDan Williams Randomization of the page allocator improves the average 1911e900a918SDan Williams utilization of a direct-mapped memory-side-cache. See section 1912e900a918SDan Williams 5.2.27 Heterogeneous Memory Attribute Table (HMAT) in the ACPI 1913e900a918SDan Williams 6.2a specification for an example of how a platform advertises 1914e900a918SDan Williams the presence of a memory-side-cache. There are also incidental 1915e900a918SDan Williams security benefits as it reduces the predictability of page 1916e900a918SDan Williams allocations to compliment SLAB_FREELIST_RANDOM, but the 1917e900a918SDan Williams default granularity of shuffling on the "MAX_ORDER - 1" i.e, 1918e900a918SDan Williams 10th order of pages is selected based on cache utilization 1919e900a918SDan Williams benefits on x86. 1920e900a918SDan Williams 1921e900a918SDan Williams While the randomization improves cache utilization it may 1922e900a918SDan Williams negatively impact workloads on platforms without a cache. For 1923e900a918SDan Williams this reason, by default, the randomization is enabled only 1924e900a918SDan Williams after runtime detection of a direct-mapped memory-side-cache. 1925e900a918SDan Williams Otherwise, the randomization may be force enabled with the 1926e900a918SDan Williams 'page_alloc.shuffle' kernel command line parameter. 1927e900a918SDan Williams 1928e900a918SDan Williams Say Y if unsure. 1929e900a918SDan Williams 1930345c905dSJoonsoo Kimconfig SLUB_CPU_PARTIAL 1931345c905dSJoonsoo Kim default y 1932b39ffbf8SUwe Kleine-König depends on SLUB && SMP 1933345c905dSJoonsoo Kim bool "SLUB per cpu partial cache" 1934345c905dSJoonsoo Kim help 193592bae787SKees Cook Per cpu partial caches accelerate objects allocation and freeing 1936345c905dSJoonsoo Kim that is local to a processor at the price of more indeterminism 1937345c905dSJoonsoo Kim in the latency of the free. On overflow these caches will be cleared 1938345c905dSJoonsoo Kim which requires the taking of locks that may cause latency spikes. 1939345c905dSJoonsoo Kim Typically one would choose no for a realtime system. 1940345c905dSJoonsoo Kim 1941ea637639SJie Zhangconfig MMAP_ALLOW_UNINITIALIZED 1942ea637639SJie Zhang bool "Allow mmapped anonymous memory to be uninitialized" 19436a108a14SDavid Rientjes depends on EXPERT && !MMU 1944ea637639SJie Zhang default n 1945ea637639SJie Zhang help 1946ea637639SJie Zhang Normally, and according to the Linux spec, anonymous memory obtained 19473903bf94SRandy Dunlap from mmap() has its contents cleared before it is passed to 1948ea637639SJie Zhang userspace. Enabling this config option allows you to request that 1949ea637639SJie Zhang mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus 1950ea637639SJie Zhang providing a huge performance boost. If this option is not enabled, 1951ea637639SJie Zhang then the flag will be ignored. 1952ea637639SJie Zhang 1953ea637639SJie Zhang This is taken advantage of by uClibc's malloc(), and also by 1954ea637639SJie Zhang ELF-FDPIC binfmt's brk and stack allocator. 1955ea637639SJie Zhang 1956ea637639SJie Zhang Because of the obvious security issues, this option should only be 1957ea637639SJie Zhang enabled on embedded devices where you control what is run in 1958ea637639SJie Zhang userspace. Since that isn't generally a problem on no-MMU systems, 1959ea637639SJie Zhang it is normally safe to say Y here. 1960ea637639SJie Zhang 1961ea637639SJie Zhang See Documentation/nommu-mmap.txt for more information. 1962ea637639SJie Zhang 1963091f6e26SDavid Howellsconfig SYSTEM_DATA_VERIFICATION 1964091f6e26SDavid Howells def_bool n 1965091f6e26SDavid Howells select SYSTEM_TRUSTED_KEYRING 1966091f6e26SDavid Howells select KEYS 1967091f6e26SDavid Howells select CRYPTO 1968d43de6c7SDavid Howells select CRYPTO_RSA 1969091f6e26SDavid Howells select ASYMMETRIC_KEY_TYPE 1970091f6e26SDavid Howells select ASYMMETRIC_PUBLIC_KEY_SUBTYPE 1971091f6e26SDavid Howells select ASN1 1972091f6e26SDavid Howells select OID_REGISTRY 1973091f6e26SDavid Howells select X509_CERTIFICATE_PARSER 1974091f6e26SDavid Howells select PKCS7_MESSAGE_PARSER 197582c04ff8SPeter Foley help 1976091f6e26SDavid Howells Provide PKCS#7 message verification using the contents of the system 1977091f6e26SDavid Howells trusted keyring to provide public keys. This then can be used for 1978091f6e26SDavid Howells module verification, kexec image verification and firmware blob 1979091f6e26SDavid Howells verification. 198082c04ff8SPeter Foley 1981125e5645SMathieu Desnoyersconfig PROFILING 1982b309a294SRobert Richter bool "Profiling support" 1983125e5645SMathieu Desnoyers help 1984125e5645SMathieu Desnoyers Say Y here to enable the extended profiling support mechanisms used 1985125e5645SMathieu Desnoyers by profilers such as OProfile. 1986125e5645SMathieu Desnoyers 19875f87f112SIngo Molnar# 19885f87f112SIngo Molnar# Place an empty function call at each tracepoint site. Can be 19895f87f112SIngo Molnar# dynamically changed for a probe function. 19905f87f112SIngo Molnar# 199197e1c18eSMathieu Desnoyersconfig TRACEPOINTS 19925f87f112SIngo Molnar bool 199397e1c18eSMathieu Desnoyers 19941da177e4SLinus Torvaldsendmenu # General setup 19951da177e4SLinus Torvalds 19961572497cSChristoph Hellwigsource "arch/Kconfig" 19971572497cSChristoph Hellwig 1998ae81f9e3SChuck Ebbertconfig RT_MUTEXES 19996341e62bSChristoph Jaeger bool 2000ae81f9e3SChuck Ebbert 20011da177e4SLinus Torvaldsconfig BASE_SMALL 20021da177e4SLinus Torvalds int 20031da177e4SLinus Torvalds default 0 if BASE_FULL 20041da177e4SLinus Torvalds default 1 if !BASE_FULL 20051da177e4SLinus Torvalds 2006c8424e77SThiago Jung Bauermannconfig MODULE_SIG_FORMAT 2007c8424e77SThiago Jung Bauermann def_bool n 2008c8424e77SThiago Jung Bauermann select SYSTEM_DATA_VERIFICATION 2009c8424e77SThiago Jung Bauermann 201066da5733SJan Engelhardtmenuconfig MODULES 20111da177e4SLinus Torvalds bool "Enable loadable module support" 201211097a03SYann E. MORIN option modules 20131da177e4SLinus Torvalds help 20141da177e4SLinus Torvalds Kernel modules are small pieces of compiled code which can 20151da177e4SLinus Torvalds be inserted in the running kernel, rather than being 20161da177e4SLinus Torvalds permanently built into the kernel. You use the "modprobe" 20171da177e4SLinus Torvalds tool to add (and sometimes remove) them. If you say Y here, 20181da177e4SLinus Torvalds many parts of the kernel can be built as modules (by 20191da177e4SLinus Torvalds answering M instead of Y where indicated): this is most 20201da177e4SLinus Torvalds useful for infrequently used options which are not required 20211da177e4SLinus Torvalds for booting. For more information, see the man pages for 20221da177e4SLinus Torvalds modprobe, lsmod, modinfo, insmod and rmmod. 20231da177e4SLinus Torvalds 20241da177e4SLinus Torvalds If you say Y here, you will need to run "make 20251da177e4SLinus Torvalds modules_install" to put the modules under /lib/modules/ 20261da177e4SLinus Torvalds where modprobe can find them (you may need to be root to do 20271da177e4SLinus Torvalds this). 20281da177e4SLinus Torvalds 20291da177e4SLinus Torvalds If unsure, say Y. 20301da177e4SLinus Torvalds 20310b0de144SRobert P. J. Dayif MODULES 20320b0de144SRobert P. J. Day 2033826e4506SLinus Torvaldsconfig MODULE_FORCE_LOAD 2034826e4506SLinus Torvalds bool "Forced module loading" 2035826e4506SLinus Torvalds default n 2036826e4506SLinus Torvalds help 203791e37a79SRusty Russell Allow loading of modules without version information (ie. modprobe 203891e37a79SRusty Russell --force). Forced module loading sets the 'F' (forced) taint flag and 203991e37a79SRusty Russell is usually a really bad idea. 2040826e4506SLinus Torvalds 20411da177e4SLinus Torvaldsconfig MODULE_UNLOAD 20421da177e4SLinus Torvalds bool "Module unloading" 20431da177e4SLinus Torvalds help 20441da177e4SLinus Torvalds Without this option you will not be able to unload any 20451da177e4SLinus Torvalds modules (note that some modules may not be unloadable 2046f7f5b675SDenys Vlasenko anyway), which makes your kernel smaller, faster 2047f7f5b675SDenys Vlasenko and simpler. If unsure, say Y. 20481da177e4SLinus Torvalds 20491da177e4SLinus Torvaldsconfig MODULE_FORCE_UNLOAD 20501da177e4SLinus Torvalds bool "Forced module unloading" 205119c92399SKees Cook depends on MODULE_UNLOAD 20521da177e4SLinus Torvalds help 20531da177e4SLinus Torvalds This option allows you to force a module to unload, even if the 20541da177e4SLinus Torvalds kernel believes it is unsafe: the kernel will remove the module 20551da177e4SLinus Torvalds without waiting for anyone to stop using it (using the -f option to 20561da177e4SLinus Torvalds rmmod). This is mainly for kernel developers and desperate users. 20571da177e4SLinus Torvalds If unsure, say N. 20581da177e4SLinus Torvalds 20591da177e4SLinus Torvaldsconfig MODVERSIONS 20600d541643SSam Ravnborg bool "Module versioning support" 20611da177e4SLinus Torvalds help 20621da177e4SLinus Torvalds Usually, you have to use modules compiled with your kernel. 20631da177e4SLinus Torvalds Saying Y here makes it sometimes possible to use modules 20641da177e4SLinus Torvalds compiled for different kernels, by adding enough information 20651da177e4SLinus Torvalds to the modules to (hopefully) spot any changes which would 20661da177e4SLinus Torvalds make them incompatible with the kernel you are running. If 20671da177e4SLinus Torvalds unsure, say N. 20681da177e4SLinus Torvalds 20692ff2b7ecSMasahiro Yamadaconfig ASM_MODVERSIONS 20702ff2b7ecSMasahiro Yamada bool 20712ff2b7ecSMasahiro Yamada default HAVE_ASM_MODVERSIONS && MODVERSIONS 20722ff2b7ecSMasahiro Yamada help 20732ff2b7ecSMasahiro Yamada This enables module versioning for exported symbols also from 20742ff2b7ecSMasahiro Yamada assembly. This can be enabled only when the target architecture 20752ff2b7ecSMasahiro Yamada supports it. 20762ff2b7ecSMasahiro Yamada 207756067812SArd Biesheuvelconfig MODULE_REL_CRCS 207856067812SArd Biesheuvel bool 207956067812SArd Biesheuvel depends on MODVERSIONS 208056067812SArd Biesheuvel 20811da177e4SLinus Torvaldsconfig MODULE_SRCVERSION_ALL 20821da177e4SLinus Torvalds bool "Source checksum for all modules" 20831da177e4SLinus Torvalds help 20841da177e4SLinus Torvalds Modules which contain a MODULE_VERSION get an extra "srcversion" 20851da177e4SLinus Torvalds field inserted into their modinfo section, which contains a 20861da177e4SLinus Torvalds sum of the source files which made it. This helps maintainers 20871da177e4SLinus Torvalds see exactly which source was used to build a module (since 20881da177e4SLinus Torvalds others sometimes change the module source without updating 20891da177e4SLinus Torvalds the version). With this option, such a "srcversion" field 20901da177e4SLinus Torvalds will be created for all modules. If unsure, say N. 20911da177e4SLinus Torvalds 2092106a4ee2SRusty Russellconfig MODULE_SIG 2093106a4ee2SRusty Russell bool "Module signature verification" 2094c8424e77SThiago Jung Bauermann select MODULE_SIG_FORMAT 2095106a4ee2SRusty Russell help 2096106a4ee2SRusty Russell Check modules for valid signatures upon load: the signature 2097106a4ee2SRusty Russell is simply appended to the module. For more information see 2098cbdc8217SNathan Chancellor <file:Documentation/admin-guide/module-signing.rst>. 2099106a4ee2SRusty Russell 2100228c37ffSDavid Howells Note that this option adds the OpenSSL development packages as a 2101228c37ffSDavid Howells kernel build dependency so that the signing tool can use its crypto 2102228c37ffSDavid Howells library. 2103228c37ffSDavid Howells 210449fcf732SDavid Howells You should enable this option if you wish to use either 210549fcf732SDavid Howells CONFIG_SECURITY_LOCKDOWN_LSM or lockdown functionality imposed via 210649fcf732SDavid Howells another LSM - otherwise unsigned modules will be loadable regardless 210749fcf732SDavid Howells of the lockdown policy. 210849fcf732SDavid Howells 2109ea0b6dcfSDavid Howells !!!WARNING!!! If you enable this option, you MUST make sure that the 2110ea0b6dcfSDavid Howells module DOES NOT get stripped after being signed. This includes the 2111ea0b6dcfSDavid Howells debuginfo strip done by some packagers (such as rpmbuild) and 2112ea0b6dcfSDavid Howells inclusion into an initramfs that wants the module size reduced. 2113ea0b6dcfSDavid Howells 2114106a4ee2SRusty Russellconfig MODULE_SIG_FORCE 2115106a4ee2SRusty Russell bool "Require modules to be validly signed" 2116106a4ee2SRusty Russell depends on MODULE_SIG 2117106a4ee2SRusty Russell help 2118106a4ee2SRusty Russell Reject unsigned modules or signed modules for which we don't have a 2119106a4ee2SRusty Russell key. Without this, such modules will simply taint the kernel. 2120ea0b6dcfSDavid Howells 2121d9d8d7edSMichal Marekconfig MODULE_SIG_ALL 2122d9d8d7edSMichal Marek bool "Automatically sign all modules" 2123d9d8d7edSMichal Marek default y 2124d9d8d7edSMichal Marek depends on MODULE_SIG 2125d9d8d7edSMichal Marek help 2126d9d8d7edSMichal Marek Sign all modules during make modules_install. Without this option, 2127d9d8d7edSMichal Marek modules must be signed manually, using the scripts/sign-file tool. 2128d9d8d7edSMichal Marek 2129d9d8d7edSMichal Marekcomment "Do not forget to sign required modules with scripts/sign-file" 2130d9d8d7edSMichal Marek depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL 2131d9d8d7edSMichal Marek 2132ea0b6dcfSDavid Howellschoice 2133ea0b6dcfSDavid Howells prompt "Which hash algorithm should modules be signed with?" 2134ea0b6dcfSDavid Howells depends on MODULE_SIG 2135ea0b6dcfSDavid Howells help 2136ea0b6dcfSDavid Howells This determines which sort of hashing algorithm will be used during 2137ea0b6dcfSDavid Howells signature generation. This algorithm _must_ be built into the kernel 2138ea0b6dcfSDavid Howells directly so that signature verification can take place. It is not 2139ea0b6dcfSDavid Howells possible to load a signed module containing the algorithm to check 2140ea0b6dcfSDavid Howells the signature on that module. 2141ea0b6dcfSDavid Howells 2142ea0b6dcfSDavid Howellsconfig MODULE_SIG_SHA1 2143ea0b6dcfSDavid Howells bool "Sign modules with SHA-1" 2144ea0b6dcfSDavid Howells select CRYPTO_SHA1 2145ea0b6dcfSDavid Howells 2146ea0b6dcfSDavid Howellsconfig MODULE_SIG_SHA224 2147ea0b6dcfSDavid Howells bool "Sign modules with SHA-224" 2148ea0b6dcfSDavid Howells select CRYPTO_SHA256 2149ea0b6dcfSDavid Howells 2150ea0b6dcfSDavid Howellsconfig MODULE_SIG_SHA256 2151ea0b6dcfSDavid Howells bool "Sign modules with SHA-256" 2152ea0b6dcfSDavid Howells select CRYPTO_SHA256 2153ea0b6dcfSDavid Howells 2154ea0b6dcfSDavid Howellsconfig MODULE_SIG_SHA384 2155ea0b6dcfSDavid Howells bool "Sign modules with SHA-384" 2156ea0b6dcfSDavid Howells select CRYPTO_SHA512 2157ea0b6dcfSDavid Howells 2158ea0b6dcfSDavid Howellsconfig MODULE_SIG_SHA512 2159ea0b6dcfSDavid Howells bool "Sign modules with SHA-512" 2160ea0b6dcfSDavid Howells select CRYPTO_SHA512 2161ea0b6dcfSDavid Howells 2162ea0b6dcfSDavid Howellsendchoice 2163ea0b6dcfSDavid Howells 216422753674SMichal Marekconfig MODULE_SIG_HASH 216522753674SMichal Marek string 216622753674SMichal Marek depends on MODULE_SIG 216722753674SMichal Marek default "sha1" if MODULE_SIG_SHA1 216822753674SMichal Marek default "sha224" if MODULE_SIG_SHA224 216922753674SMichal Marek default "sha256" if MODULE_SIG_SHA256 217022753674SMichal Marek default "sha384" if MODULE_SIG_SHA384 217122753674SMichal Marek default "sha512" if MODULE_SIG_SHA512 217222753674SMichal Marek 2173beb50df3SBertrand Jacquinconfig MODULE_COMPRESS 2174beb50df3SBertrand Jacquin bool "Compress modules on installation" 2175beb50df3SBertrand Jacquin help 2176beb50df3SBertrand Jacquin 2177b6c09b51SRusty Russell Compresses kernel modules when 'make modules_install' is run; gzip or 2178b6c09b51SRusty Russell xz depending on "Compression algorithm" below. 2179beb50df3SBertrand Jacquin 2180b6c09b51SRusty Russell module-init-tools MAY support gzip, and kmod MAY support gzip and xz. 2181beb50df3SBertrand Jacquin 2182b6c09b51SRusty Russell Out-of-tree kernel modules installed using Kbuild will also be 2183b6c09b51SRusty Russell compressed upon installation. 2184beb50df3SBertrand Jacquin 2185b6c09b51SRusty Russell Note: for modules inside an initrd or initramfs, it's more efficient 2186b6c09b51SRusty Russell to compress the whole initrd or initramfs instead. 2187beb50df3SBertrand Jacquin 2188b6c09b51SRusty Russell Note: This is fully compatible with signed modules. 2189b6c09b51SRusty Russell 2190b6c09b51SRusty Russell If in doubt, say N. 2191beb50df3SBertrand Jacquin 2192beb50df3SBertrand Jacquinchoice 2193beb50df3SBertrand Jacquin prompt "Compression algorithm" 2194beb50df3SBertrand Jacquin depends on MODULE_COMPRESS 2195beb50df3SBertrand Jacquin default MODULE_COMPRESS_GZIP 2196beb50df3SBertrand Jacquin help 2197beb50df3SBertrand Jacquin This determines which sort of compression will be used during 2198beb50df3SBertrand Jacquin 'make modules_install'. 2199beb50df3SBertrand Jacquin 2200beb50df3SBertrand Jacquin GZIP (default) and XZ are supported. 2201beb50df3SBertrand Jacquin 2202beb50df3SBertrand Jacquinconfig MODULE_COMPRESS_GZIP 2203beb50df3SBertrand Jacquin bool "GZIP" 2204beb50df3SBertrand Jacquin 2205beb50df3SBertrand Jacquinconfig MODULE_COMPRESS_XZ 2206beb50df3SBertrand Jacquin bool "XZ" 2207beb50df3SBertrand Jacquin 2208beb50df3SBertrand Jacquinendchoice 2209beb50df3SBertrand Jacquin 22103d52ec5eSMatthias Maennichconfig MODULE_ALLOW_MISSING_NAMESPACE_IMPORTS 22113d52ec5eSMatthias Maennich bool "Allow loading of modules with missing namespace imports" 22123d52ec5eSMatthias Maennich help 22133d52ec5eSMatthias Maennich Symbols exported with EXPORT_SYMBOL_NS*() are considered exported in 22143d52ec5eSMatthias Maennich a namespace. A module that makes use of a symbol exported with such a 22153d52ec5eSMatthias Maennich namespace is required to import the namespace via MODULE_IMPORT_NS(). 22163d52ec5eSMatthias Maennich There is no technical reason to enforce correct namespace imports, 22173d52ec5eSMatthias Maennich but it creates consistency between symbols defining namespaces and 22183d52ec5eSMatthias Maennich users importing namespaces they make use of. This option relaxes this 22193d52ec5eSMatthias Maennich requirement and lifts the enforcement when loading a module. 22203d52ec5eSMatthias Maennich 22213d52ec5eSMatthias Maennich If unsure, say N. 22223d52ec5eSMatthias Maennich 2223efd9763dSMasahiro Yamadaconfig UNUSED_SYMBOLS 2224efd9763dSMasahiro Yamada bool "Enable unused/obsolete exported symbols" 2225efd9763dSMasahiro Yamada default y if X86 2226efd9763dSMasahiro Yamada help 2227efd9763dSMasahiro Yamada Unused but exported symbols make the kernel needlessly bigger. For 2228efd9763dSMasahiro Yamada that reason most of these unused exports will soon be removed. This 2229efd9763dSMasahiro Yamada option is provided temporarily to provide a transition period in case 2230efd9763dSMasahiro Yamada some external kernel module needs one of these symbols anyway. If you 2231efd9763dSMasahiro Yamada encounter such a case in your module, consider if you are actually 2232efd9763dSMasahiro Yamada using the right API. (rationale: since nobody in the kernel is using 2233efd9763dSMasahiro Yamada this in a module, there is a pretty good chance it's actually the 2234efd9763dSMasahiro Yamada wrong interface to use). If you really need the symbol, please send a 2235efd9763dSMasahiro Yamada mail to the linux kernel mailing list mentioning the symbol and why 2236efd9763dSMasahiro Yamada you really need it, and what the merge plan to the mainline kernel for 2237efd9763dSMasahiro Yamada your module is. 2238efd9763dSMasahiro Yamada 2239dbacb0efSNicolas Pitreconfig TRIM_UNUSED_KSYMS 2240dbacb0efSNicolas Pitre bool "Trim unused exported kernel symbols" 2241d189c2a4SMasahiro Yamada depends on !UNUSED_SYMBOLS 2242dbacb0efSNicolas Pitre help 2243dbacb0efSNicolas Pitre The kernel and some modules make many symbols available for 2244dbacb0efSNicolas Pitre other modules to use via EXPORT_SYMBOL() and variants. Depending 2245dbacb0efSNicolas Pitre on the set of modules being selected in your kernel configuration, 2246dbacb0efSNicolas Pitre many of those exported symbols might never be used. 2247dbacb0efSNicolas Pitre 2248dbacb0efSNicolas Pitre This option allows for unused exported symbols to be dropped from 2249dbacb0efSNicolas Pitre the build. In turn, this provides the compiler more opportunities 2250dbacb0efSNicolas Pitre (especially when using LTO) for optimizing the code and reducing 2251dbacb0efSNicolas Pitre binary size. This might have some security advantages as well. 2252dbacb0efSNicolas Pitre 2253f1cb637eSValdis Kletnieks If unsure, or if you need to build out-of-tree modules, say N. 2254dbacb0efSNicolas Pitre 22551518c633SQuentin Perretconfig UNUSED_KSYMS_WHITELIST 22561518c633SQuentin Perret string "Whitelist of symbols to keep in ksymtab" 22571518c633SQuentin Perret depends on TRIM_UNUSED_KSYMS 22581518c633SQuentin Perret help 22591518c633SQuentin Perret By default, all unused exported symbols will be un-exported from the 22601518c633SQuentin Perret build when TRIM_UNUSED_KSYMS is selected. 22611518c633SQuentin Perret 22621518c633SQuentin Perret UNUSED_KSYMS_WHITELIST allows to whitelist symbols that must be kept 22631518c633SQuentin Perret exported at all times, even in absence of in-tree users. The value to 22641518c633SQuentin Perret set here is the path to a text file containing the list of symbols, 22651518c633SQuentin Perret one per line. The path can be absolute, or relative to the kernel 22661518c633SQuentin Perret source tree. 22671518c633SQuentin Perret 22680b0de144SRobert P. J. Dayendif # MODULES 22690b0de144SRobert P. J. Day 22706c9692e2SPeter Zijlstraconfig MODULES_TREE_LOOKUP 22716c9692e2SPeter Zijlstra def_bool y 22726c9692e2SPeter Zijlstra depends on PERF_EVENTS || TRACING 22736c9692e2SPeter Zijlstra 227498a79d6aSRusty Russellconfig INIT_ALL_POSSIBLE 227598a79d6aSRusty Russell bool 227698a79d6aSRusty Russell help 22775f054e31SRusty Russell Back when each arch used to define their own cpu_online_mask and 22785f054e31SRusty Russell cpu_possible_mask, some of them chose to initialize cpu_possible_mask 227998a79d6aSRusty Russell with all 1s, and others with all 0s. When they were centralised, 228098a79d6aSRusty Russell it was better to provide this option than to break all the archs 2281692105b8SMatt LaPlante and have several arch maintainers pursuing me down dark alleys. 228298a79d6aSRusty Russell 22833a65dfe8SJens Axboesource "block/Kconfig" 2284e98c3202SAvi Kivity 2285e98c3202SAvi Kivityconfig PREEMPT_NOTIFIERS 2286e98c3202SAvi Kivity bool 2287e260be67SPaul E. McKenney 228816295becSSteffen Klassertconfig PADATA 228916295becSSteffen Klassert depends on SMP 229016295becSSteffen Klassert bool 229116295becSSteffen Klassert 22924520c6a4SDavid Howellsconfig ASN1 22934520c6a4SDavid Howells tristate 22944520c6a4SDavid Howells help 22954520c6a4SDavid Howells Build a simple ASN.1 grammar compiler that produces a bytecode output 22964520c6a4SDavid Howells that can be interpreted by the ASN.1 stream decoder and used to 22974520c6a4SDavid Howells inform it as to what tags are to be expected in a stream and what 22984520c6a4SDavid Howells functions to call on what tags. 22994520c6a4SDavid Howells 23006beb0009SThomas Gleixnersource "kernel/Kconfig.locks" 2301e61938a9SMathieu Desnoyers 23020ebeea8cSDaniel Borkmannconfig ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE 23030ebeea8cSDaniel Borkmann bool 23040ebeea8cSDaniel Borkmann 2305e61938a9SMathieu Desnoyersconfig ARCH_HAS_SYNC_CORE_BEFORE_USERMODE 2306e61938a9SMathieu Desnoyers bool 23071bd21c6cSDominik Brodowski 23081bd21c6cSDominik Brodowski# It may be useful for an architecture to override the definitions of the 23097303e30eSDominik Brodowski# SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h> 23107303e30eSDominik Brodowski# and the COMPAT_ variants in <linux/compat.h>, in particular to use a 23117303e30eSDominik Brodowski# different calling convention for syscalls. They can also override the 23127303e30eSDominik Brodowski# macros for not-implemented syscalls in kernel/sys_ni.c and 23137303e30eSDominik Brodowski# kernel/time/posix-stubs.c. All these overrides need to be available in 23147303e30eSDominik Brodowski# <asm/syscall_wrapper.h>. 23151bd21c6cSDominik Brodowskiconfig ARCH_HAS_SYSCALL_WRAPPER 23161bd21c6cSDominik Brodowski def_bool n 2317