1# SPDX-License-Identifier: GPL-2.0-only 2config CC_VERSION_TEXT 3 string 4 default "$(CC_VERSION_TEXT)" 5 help 6 This is used in unclear ways: 7 8 - Re-run Kconfig when the compiler is updated 9 The 'default' property references the environment variable, 10 CC_VERSION_TEXT so it is recorded in include/config/auto.conf.cmd. 11 When the compiler is updated, Kconfig will be invoked. 12 13 - Ensure full rebuild when the compiler is updated 14 include/linux/compiler-version.h contains this option in the comment 15 line so fixdep adds include/config/CC_VERSION_TEXT into the 16 auto-generated dependency. When the compiler is updated, syncconfig 17 will touch it and then every file will be rebuilt. 18 19config CC_IS_GCC 20 def_bool $(success,test "$(cc-name)" = GCC) 21 22config GCC_VERSION 23 int 24 default $(cc-version) if CC_IS_GCC 25 default 0 26 27config CC_IS_CLANG 28 def_bool $(success,test "$(cc-name)" = Clang) 29 30config CLANG_VERSION 31 int 32 default $(cc-version) if CC_IS_CLANG 33 default 0 34 35config AS_IS_GNU 36 def_bool $(success,test "$(as-name)" = GNU) 37 38config AS_IS_LLVM 39 def_bool $(success,test "$(as-name)" = LLVM) 40 41config AS_VERSION 42 int 43 # Use clang version if this is the integrated assembler 44 default CLANG_VERSION if AS_IS_LLVM 45 default $(as-version) 46 47config LD_IS_BFD 48 def_bool $(success,test "$(ld-name)" = BFD) 49 50config LD_VERSION 51 int 52 default $(ld-version) if LD_IS_BFD 53 default 0 54 55config LD_IS_LLD 56 def_bool $(success,test "$(ld-name)" = LLD) 57 58config LLD_VERSION 59 int 60 default $(ld-version) if LD_IS_LLD 61 default 0 62 63config RUST_IS_AVAILABLE 64 def_bool $(success,$(srctree)/scripts/rust_is_available.sh) 65 help 66 This shows whether a suitable Rust toolchain is available (found). 67 68 Please see Documentation/rust/quick-start.rst for instructions on how 69 to satisfy the build requirements of Rust support. 70 71 In particular, the Makefile target 'rustavailable' is useful to check 72 why the Rust toolchain is not being detected. 73 74config CC_CAN_LINK 75 bool 76 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag)) if 64BIT 77 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag)) 78 79config CC_CAN_LINK_STATIC 80 bool 81 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag) -static) if 64BIT 82 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag) -static) 83 84config CC_HAS_ASM_GOTO_OUTPUT 85 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) 86 87config CC_HAS_ASM_GOTO_TIED_OUTPUT 88 depends on CC_HAS_ASM_GOTO_OUTPUT 89 # Detect buggy gcc and clang, fixed in gcc-11 clang-14. 90 def_bool $(success,echo 'int foo(int *x) { asm goto (".long (%l[bar]) - .": "+m"(*x) ::: bar); return *x; bar: return 0; }' | $CC -x c - -c -o /dev/null) 91 92config TOOLS_SUPPORT_RELR 93 def_bool $(success,env "CC=$(CC)" "LD=$(LD)" "NM=$(NM)" "OBJCOPY=$(OBJCOPY)" $(srctree)/scripts/tools-support-relr.sh) 94 95config CC_HAS_ASM_INLINE 96 def_bool $(success,echo 'void foo(void) { asm inline (""); }' | $(CC) -x c - -c -o /dev/null) 97 98config CC_HAS_NO_PROFILE_FN_ATTR 99 def_bool $(success,echo '__attribute__((no_profile_instrument_function)) int x();' | $(CC) -x c - -c -o /dev/null -Werror) 100 101config PAHOLE_VERSION 102 int 103 default $(shell,$(srctree)/scripts/pahole-version.sh $(PAHOLE)) 104 105config CONSTRUCTORS 106 bool 107 108config IRQ_WORK 109 bool 110 111config BUILDTIME_TABLE_SORT 112 bool 113 114config THREAD_INFO_IN_TASK 115 bool 116 help 117 Select this to move thread_info off the stack into task_struct. To 118 make this work, an arch will need to remove all thread_info fields 119 except flags and fix any runtime bugs. 120 121 One subtle change that will be needed is to use try_get_task_stack() 122 and put_task_stack() in save_thread_stack_tsk() and get_wchan(). 123 124menu "General setup" 125 126config BROKEN 127 bool 128 129config BROKEN_ON_SMP 130 bool 131 depends on BROKEN || !SMP 132 default y 133 134config INIT_ENV_ARG_LIMIT 135 int 136 default 32 if !UML 137 default 128 if UML 138 help 139 Maximum of each of the number of arguments and environment 140 variables passed to init from the kernel command line. 141 142config COMPILE_TEST 143 bool "Compile also drivers which will not load" 144 depends on HAS_IOMEM 145 help 146 Some drivers can be compiled on a different platform than they are 147 intended to be run on. Despite they cannot be loaded there (or even 148 when they load they cannot be used due to missing HW support), 149 developers still, opposing to distributors, might want to build such 150 drivers to compile-test them. 151 152 If you are a developer and want to build everything available, say Y 153 here. If you are a user/distributor, say N here to exclude useless 154 drivers to be distributed. 155 156config WERROR 157 bool "Compile the kernel with warnings as errors" 158 default COMPILE_TEST 159 help 160 A kernel build should not cause any compiler warnings, and this 161 enables the '-Werror' (for C) and '-Dwarnings' (for Rust) flags 162 to enforce that rule by default. 163 164 However, if you have a new (or very old) compiler with odd and 165 unusual warnings, or you have some architecture with problems, 166 you may need to disable this config option in order to 167 successfully build the kernel. 168 169 If in doubt, say Y. 170 171config UAPI_HEADER_TEST 172 bool "Compile test UAPI headers" 173 depends on HEADERS_INSTALL && CC_CAN_LINK 174 help 175 Compile test headers exported to user-space to ensure they are 176 self-contained, i.e. compilable as standalone units. 177 178 If you are a developer or tester and want to ensure the exported 179 headers are self-contained, say Y here. Otherwise, choose N. 180 181config LOCALVERSION 182 string "Local version - append to kernel release" 183 help 184 Append an extra string to the end of your kernel version. 185 This will show up when you type uname, for example. 186 The string you set here will be appended after the contents of 187 any files with a filename matching localversion* in your 188 object and source tree, in that order. Your total string can 189 be a maximum of 64 characters. 190 191config LOCALVERSION_AUTO 192 bool "Automatically append version information to the version string" 193 default y 194 depends on !COMPILE_TEST 195 help 196 This will try to automatically determine if the current tree is a 197 release tree by looking for git tags that belong to the current 198 top of tree revision. 199 200 A string of the format -gxxxxxxxx will be added to the localversion 201 if a git-based tree is found. The string generated by this will be 202 appended after any matching localversion* files, and after the value 203 set in CONFIG_LOCALVERSION. 204 205 (The actual string used here is the first eight characters produced 206 by running the command: 207 208 $ git rev-parse --verify HEAD 209 210 which is done within the script "scripts/setlocalversion".) 211 212config BUILD_SALT 213 string "Build ID Salt" 214 default "" 215 help 216 The build ID is used to link binaries and their debug info. Setting 217 this option will use the value in the calculation of the build id. 218 This is mostly useful for distributions which want to ensure the 219 build is unique between builds. It's safe to leave the default. 220 221config HAVE_KERNEL_GZIP 222 bool 223 224config HAVE_KERNEL_BZIP2 225 bool 226 227config HAVE_KERNEL_LZMA 228 bool 229 230config HAVE_KERNEL_XZ 231 bool 232 233config HAVE_KERNEL_LZO 234 bool 235 236config HAVE_KERNEL_LZ4 237 bool 238 239config HAVE_KERNEL_ZSTD 240 bool 241 242config HAVE_KERNEL_UNCOMPRESSED 243 bool 244 245choice 246 prompt "Kernel compression mode" 247 default KERNEL_GZIP 248 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_ZSTD || HAVE_KERNEL_UNCOMPRESSED 249 help 250 The linux kernel is a kind of self-extracting executable. 251 Several compression algorithms are available, which differ 252 in efficiency, compression and decompression speed. 253 Compression speed is only relevant when building a kernel. 254 Decompression speed is relevant at each boot. 255 256 If you have any problems with bzip2 or lzma compressed 257 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 258 version of this functionality (bzip2 only), for 2.4, was 259 supplied by Christian Ludwig) 260 261 High compression options are mostly useful for users, who 262 are low on disk space (embedded systems), but for whom ram 263 size matters less. 264 265 If in doubt, select 'gzip' 266 267config KERNEL_GZIP 268 bool "Gzip" 269 depends on HAVE_KERNEL_GZIP 270 help 271 The old and tried gzip compression. It provides a good balance 272 between compression ratio and decompression speed. 273 274config KERNEL_BZIP2 275 bool "Bzip2" 276 depends on HAVE_KERNEL_BZIP2 277 help 278 Its compression ratio and speed is intermediate. 279 Decompression speed is slowest among the choices. The kernel 280 size is about 10% smaller with bzip2, in comparison to gzip. 281 Bzip2 uses a large amount of memory. For modern kernels you 282 will need at least 8MB RAM or more for booting. 283 284config KERNEL_LZMA 285 bool "LZMA" 286 depends on HAVE_KERNEL_LZMA 287 help 288 This compression algorithm's ratio is best. Decompression speed 289 is between gzip and bzip2. Compression is slowest. 290 The kernel size is about 33% smaller with LZMA in comparison to gzip. 291 292config KERNEL_XZ 293 bool "XZ" 294 depends on HAVE_KERNEL_XZ 295 help 296 XZ uses the LZMA2 algorithm and instruction set specific 297 BCJ filters which can improve compression ratio of executable 298 code. The size of the kernel is about 30% smaller with XZ in 299 comparison to gzip. On architectures for which there is a BCJ 300 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 301 will create a few percent smaller kernel than plain LZMA. 302 303 The speed is about the same as with LZMA: The decompression 304 speed of XZ is better than that of bzip2 but worse than gzip 305 and LZO. Compression is slow. 306 307config KERNEL_LZO 308 bool "LZO" 309 depends on HAVE_KERNEL_LZO 310 help 311 Its compression ratio is the poorest among the choices. The kernel 312 size is about 10% bigger than gzip; however its speed 313 (both compression and decompression) is the fastest. 314 315config KERNEL_LZ4 316 bool "LZ4" 317 depends on HAVE_KERNEL_LZ4 318 help 319 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding. 320 A preliminary version of LZ4 de/compression tool is available at 321 <https://code.google.com/p/lz4/>. 322 323 Its compression ratio is worse than LZO. The size of the kernel 324 is about 8% bigger than LZO. But the decompression speed is 325 faster than LZO. 326 327config KERNEL_ZSTD 328 bool "ZSTD" 329 depends on HAVE_KERNEL_ZSTD 330 help 331 ZSTD is a compression algorithm targeting intermediate compression 332 with fast decompression speed. It will compress better than GZIP and 333 decompress around the same speed as LZO, but slower than LZ4. You 334 will need at least 192 KB RAM or more for booting. The zstd command 335 line tool is required for compression. 336 337config KERNEL_UNCOMPRESSED 338 bool "None" 339 depends on HAVE_KERNEL_UNCOMPRESSED 340 help 341 Produce uncompressed kernel image. This option is usually not what 342 you want. It is useful for debugging the kernel in slow simulation 343 environments, where decompressing and moving the kernel is awfully 344 slow. This option allows early boot code to skip the decompressor 345 and jump right at uncompressed kernel image. 346 347endchoice 348 349config DEFAULT_INIT 350 string "Default init path" 351 default "" 352 help 353 This option determines the default init for the system if no init= 354 option is passed on the kernel command line. If the requested path is 355 not present, we will still then move on to attempting further 356 locations (e.g. /sbin/init, etc). If this is empty, we will just use 357 the fallback list when init= is not passed. 358 359config DEFAULT_HOSTNAME 360 string "Default hostname" 361 default "(none)" 362 help 363 This option determines the default system hostname before userspace 364 calls sethostname(2). The kernel traditionally uses "(none)" here, 365 but you may wish to use a different default here to make a minimal 366 system more usable with less configuration. 367 368config SYSVIPC 369 bool "System V IPC" 370 help 371 Inter Process Communication is a suite of library functions and 372 system calls which let processes (running programs) synchronize and 373 exchange information. It is generally considered to be a good thing, 374 and some programs won't run unless you say Y here. In particular, if 375 you want to run the DOS emulator dosemu under Linux (read the 376 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 377 you'll need to say Y here. 378 379 You can find documentation about IPC with "info ipc" and also in 380 section 6.4 of the Linux Programmer's Guide, available from 381 <http://www.tldp.org/guides.html>. 382 383config SYSVIPC_SYSCTL 384 bool 385 depends on SYSVIPC 386 depends on SYSCTL 387 default y 388 389config SYSVIPC_COMPAT 390 def_bool y 391 depends on COMPAT && SYSVIPC 392 393config POSIX_MQUEUE 394 bool "POSIX Message Queues" 395 depends on NET 396 help 397 POSIX variant of message queues is a part of IPC. In POSIX message 398 queues every message has a priority which decides about succession 399 of receiving it by a process. If you want to compile and run 400 programs written e.g. for Solaris with use of its POSIX message 401 queues (functions mq_*) say Y here. 402 403 POSIX message queues are visible as a filesystem called 'mqueue' 404 and can be mounted somewhere if you want to do filesystem 405 operations on message queues. 406 407 If unsure, say Y. 408 409config POSIX_MQUEUE_SYSCTL 410 bool 411 depends on POSIX_MQUEUE 412 depends on SYSCTL 413 default y 414 415config WATCH_QUEUE 416 bool "General notification queue" 417 default n 418 help 419 420 This is a general notification queue for the kernel to pass events to 421 userspace by splicing them into pipes. It can be used in conjunction 422 with watches for key/keyring change notifications and device 423 notifications. 424 425 See Documentation/core-api/watch_queue.rst 426 427config CROSS_MEMORY_ATTACH 428 bool "Enable process_vm_readv/writev syscalls" 429 depends on MMU 430 default y 431 help 432 Enabling this option adds the system calls process_vm_readv and 433 process_vm_writev which allow a process with the correct privileges 434 to directly read from or write to another process' address space. 435 See the man page for more details. 436 437config USELIB 438 bool "uselib syscall (for libc5 and earlier)" 439 default ALPHA || M68K || SPARC 440 help 441 This option enables the uselib syscall, a system call used in the 442 dynamic linker from libc5 and earlier. glibc does not use this 443 system call. If you intend to run programs built on libc5 or 444 earlier, you may need to enable this syscall. Current systems 445 running glibc can safely disable this. 446 447config AUDIT 448 bool "Auditing support" 449 depends on NET 450 help 451 Enable auditing infrastructure that can be used with another 452 kernel subsystem, such as SELinux (which requires this for 453 logging of avc messages output). System call auditing is included 454 on architectures which support it. 455 456config HAVE_ARCH_AUDITSYSCALL 457 bool 458 459config AUDITSYSCALL 460 def_bool y 461 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL 462 select FSNOTIFY 463 464source "kernel/irq/Kconfig" 465source "kernel/time/Kconfig" 466source "kernel/bpf/Kconfig" 467source "kernel/Kconfig.preempt" 468 469menu "CPU/Task time and stats accounting" 470 471config VIRT_CPU_ACCOUNTING 472 bool 473 474choice 475 prompt "Cputime accounting" 476 default TICK_CPU_ACCOUNTING 477 478# Kind of a stub config for the pure tick based cputime accounting 479config TICK_CPU_ACCOUNTING 480 bool "Simple tick based cputime accounting" 481 depends on !S390 && !NO_HZ_FULL 482 help 483 This is the basic tick based cputime accounting that maintains 484 statistics about user, system and idle time spent on per jiffies 485 granularity. 486 487 If unsure, say Y. 488 489config VIRT_CPU_ACCOUNTING_NATIVE 490 bool "Deterministic task and CPU time accounting" 491 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL 492 select VIRT_CPU_ACCOUNTING 493 help 494 Select this option to enable more accurate task and CPU time 495 accounting. This is done by reading a CPU counter on each 496 kernel entry and exit and on transitions within the kernel 497 between system, softirq and hardirq state, so there is a 498 small performance impact. In the case of s390 or IBM POWER > 5, 499 this also enables accounting of stolen time on logically-partitioned 500 systems. 501 502config VIRT_CPU_ACCOUNTING_GEN 503 bool "Full dynticks CPU time accounting" 504 depends on HAVE_CONTEXT_TRACKING_USER 505 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN 506 depends on GENERIC_CLOCKEVENTS 507 select VIRT_CPU_ACCOUNTING 508 select CONTEXT_TRACKING_USER 509 help 510 Select this option to enable task and CPU time accounting on full 511 dynticks systems. This accounting is implemented by watching every 512 kernel-user boundaries using the context tracking subsystem. 513 The accounting is thus performed at the expense of some significant 514 overhead. 515 516 For now this is only useful if you are working on the full 517 dynticks subsystem development. 518 519 If unsure, say N. 520 521endchoice 522 523config IRQ_TIME_ACCOUNTING 524 bool "Fine granularity task level IRQ time accounting" 525 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE 526 help 527 Select this option to enable fine granularity task irq time 528 accounting. This is done by reading a timestamp on each 529 transitions between softirq and hardirq state, so there can be a 530 small performance impact. 531 532 If in doubt, say N here. 533 534config HAVE_SCHED_AVG_IRQ 535 def_bool y 536 depends on IRQ_TIME_ACCOUNTING || PARAVIRT_TIME_ACCOUNTING 537 depends on SMP 538 539config SCHED_THERMAL_PRESSURE 540 bool 541 default y if ARM && ARM_CPU_TOPOLOGY 542 default y if ARM64 543 depends on SMP 544 depends on CPU_FREQ_THERMAL 545 help 546 Select this option to enable thermal pressure accounting in the 547 scheduler. Thermal pressure is the value conveyed to the scheduler 548 that reflects the reduction in CPU compute capacity resulted from 549 thermal throttling. Thermal throttling occurs when the performance of 550 a CPU is capped due to high operating temperatures. 551 552 If selected, the scheduler will be able to balance tasks accordingly, 553 i.e. put less load on throttled CPUs than on non/less throttled ones. 554 555 This requires the architecture to implement 556 arch_update_thermal_pressure() and arch_scale_thermal_pressure(). 557 558config BSD_PROCESS_ACCT 559 bool "BSD Process Accounting" 560 depends on MULTIUSER 561 help 562 If you say Y here, a user level program will be able to instruct the 563 kernel (via a special system call) to write process accounting 564 information to a file: whenever a process exits, information about 565 that process will be appended to the file by the kernel. The 566 information includes things such as creation time, owning user, 567 command name, memory usage, controlling terminal etc. (the complete 568 list is in the struct acct in <file:include/linux/acct.h>). It is 569 up to the user level program to do useful things with this 570 information. This is generally a good idea, so say Y. 571 572config BSD_PROCESS_ACCT_V3 573 bool "BSD Process Accounting version 3 file format" 574 depends on BSD_PROCESS_ACCT 575 default n 576 help 577 If you say Y here, the process accounting information is written 578 in a new file format that also logs the process IDs of each 579 process and its parent. Note that this file format is incompatible 580 with previous v0/v1/v2 file formats, so you will need updated tools 581 for processing it. A preliminary version of these tools is available 582 at <http://www.gnu.org/software/acct/>. 583 584config TASKSTATS 585 bool "Export task/process statistics through netlink" 586 depends on NET 587 depends on MULTIUSER 588 default n 589 help 590 Export selected statistics for tasks/processes through the 591 generic netlink interface. Unlike BSD process accounting, the 592 statistics are available during the lifetime of tasks/processes as 593 responses to commands. Like BSD accounting, they are sent to user 594 space on task exit. 595 596 Say N if unsure. 597 598config TASK_DELAY_ACCT 599 bool "Enable per-task delay accounting" 600 depends on TASKSTATS 601 select SCHED_INFO 602 help 603 Collect information on time spent by a task waiting for system 604 resources like cpu, synchronous block I/O completion and swapping 605 in pages. Such statistics can help in setting a task's priorities 606 relative to other tasks for cpu, io, rss limits etc. 607 608 Say N if unsure. 609 610config TASK_XACCT 611 bool "Enable extended accounting over taskstats" 612 depends on TASKSTATS 613 help 614 Collect extended task accounting data and send the data 615 to userland for processing over the taskstats interface. 616 617 Say N if unsure. 618 619config TASK_IO_ACCOUNTING 620 bool "Enable per-task storage I/O accounting" 621 depends on TASK_XACCT 622 help 623 Collect information on the number of bytes of storage I/O which this 624 task has caused. 625 626 Say N if unsure. 627 628config PSI 629 bool "Pressure stall information tracking" 630 help 631 Collect metrics that indicate how overcommitted the CPU, memory, 632 and IO capacity are in the system. 633 634 If you say Y here, the kernel will create /proc/pressure/ with the 635 pressure statistics files cpu, memory, and io. These will indicate 636 the share of walltime in which some or all tasks in the system are 637 delayed due to contention of the respective resource. 638 639 In kernels with cgroup support, cgroups (cgroup2 only) will 640 have cpu.pressure, memory.pressure, and io.pressure files, 641 which aggregate pressure stalls for the grouped tasks only. 642 643 For more details see Documentation/accounting/psi.rst. 644 645 Say N if unsure. 646 647config PSI_DEFAULT_DISABLED 648 bool "Require boot parameter to enable pressure stall information tracking" 649 default n 650 depends on PSI 651 help 652 If set, pressure stall information tracking will be disabled 653 per default but can be enabled through passing psi=1 on the 654 kernel commandline during boot. 655 656 This feature adds some code to the task wakeup and sleep 657 paths of the scheduler. The overhead is too low to affect 658 common scheduling-intense workloads in practice (such as 659 webservers, memcache), but it does show up in artificial 660 scheduler stress tests, such as hackbench. 661 662 If you are paranoid and not sure what the kernel will be 663 used for, say Y. 664 665 Say N if unsure. 666 667endmenu # "CPU/Task time and stats accounting" 668 669config CPU_ISOLATION 670 bool "CPU isolation" 671 depends on SMP || COMPILE_TEST 672 default y 673 help 674 Make sure that CPUs running critical tasks are not disturbed by 675 any source of "noise" such as unbound workqueues, timers, kthreads... 676 Unbound jobs get offloaded to housekeeping CPUs. This is driven by 677 the "isolcpus=" boot parameter. 678 679 Say Y if unsure. 680 681source "kernel/rcu/Kconfig" 682 683config BUILD_BIN2C 684 bool 685 default n 686 687config IKCONFIG 688 tristate "Kernel .config support" 689 help 690 This option enables the complete Linux kernel ".config" file 691 contents to be saved in the kernel. It provides documentation 692 of which kernel options are used in a running kernel or in an 693 on-disk kernel. This information can be extracted from the kernel 694 image file with the script scripts/extract-ikconfig and used as 695 input to rebuild the current kernel or to build another kernel. 696 It can also be extracted from a running kernel by reading 697 /proc/config.gz if enabled (below). 698 699config IKCONFIG_PROC 700 bool "Enable access to .config through /proc/config.gz" 701 depends on IKCONFIG && PROC_FS 702 help 703 This option enables access to the kernel configuration file 704 through /proc/config.gz. 705 706config IKHEADERS 707 tristate "Enable kernel headers through /sys/kernel/kheaders.tar.xz" 708 depends on SYSFS 709 help 710 This option enables access to the in-kernel headers that are generated during 711 the build process. These can be used to build eBPF tracing programs, 712 or similar programs. If you build the headers as a module, a module called 713 kheaders.ko is built which can be loaded on-demand to get access to headers. 714 715config LOG_BUF_SHIFT 716 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 717 range 12 25 718 default 17 719 depends on PRINTK 720 help 721 Select the minimal kernel log buffer size as a power of 2. 722 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config 723 parameter, see below. Any higher size also might be forced 724 by "log_buf_len" boot parameter. 725 726 Examples: 727 17 => 128 KB 728 16 => 64 KB 729 15 => 32 KB 730 14 => 16 KB 731 13 => 8 KB 732 12 => 4 KB 733 734config LOG_CPU_MAX_BUF_SHIFT 735 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)" 736 depends on SMP 737 range 0 21 738 default 12 if !BASE_SMALL 739 default 0 if BASE_SMALL 740 depends on PRINTK 741 help 742 This option allows to increase the default ring buffer size 743 according to the number of CPUs. The value defines the contribution 744 of each CPU as a power of 2. The used space is typically only few 745 lines however it might be much more when problems are reported, 746 e.g. backtraces. 747 748 The increased size means that a new buffer has to be allocated and 749 the original static one is unused. It makes sense only on systems 750 with more CPUs. Therefore this value is used only when the sum of 751 contributions is greater than the half of the default kernel ring 752 buffer as defined by LOG_BUF_SHIFT. The default values are set 753 so that more than 16 CPUs are needed to trigger the allocation. 754 755 Also this option is ignored when "log_buf_len" kernel parameter is 756 used as it forces an exact (power of two) size of the ring buffer. 757 758 The number of possible CPUs is used for this computation ignoring 759 hotplugging making the computation optimal for the worst case 760 scenario while allowing a simple algorithm to be used from bootup. 761 762 Examples shift values and their meaning: 763 17 => 128 KB for each CPU 764 16 => 64 KB for each CPU 765 15 => 32 KB for each CPU 766 14 => 16 KB for each CPU 767 13 => 8 KB for each CPU 768 12 => 4 KB for each CPU 769 770config PRINTK_SAFE_LOG_BUF_SHIFT 771 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)" 772 range 10 21 773 default 13 774 depends on PRINTK 775 help 776 Select the size of an alternate printk per-CPU buffer where messages 777 printed from usafe contexts are temporary stored. One example would 778 be NMI messages, another one - printk recursion. The messages are 779 copied to the main log buffer in a safe context to avoid a deadlock. 780 The value defines the size as a power of 2. 781 782 Those messages are rare and limited. The largest one is when 783 a backtrace is printed. It usually fits into 4KB. Select 784 8KB if you want to be on the safe side. 785 786 Examples: 787 17 => 128 KB for each CPU 788 16 => 64 KB for each CPU 789 15 => 32 KB for each CPU 790 14 => 16 KB for each CPU 791 13 => 8 KB for each CPU 792 12 => 4 KB for each CPU 793 794config PRINTK_INDEX 795 bool "Printk indexing debugfs interface" 796 depends on PRINTK && DEBUG_FS 797 help 798 Add support for indexing of all printk formats known at compile time 799 at <debugfs>/printk/index/<module>. 800 801 This can be used as part of maintaining daemons which monitor 802 /dev/kmsg, as it permits auditing the printk formats present in a 803 kernel, allowing detection of cases where monitored printks are 804 changed or no longer present. 805 806 There is no additional runtime cost to printk with this enabled. 807 808# 809# Architectures with an unreliable sched_clock() should select this: 810# 811config HAVE_UNSTABLE_SCHED_CLOCK 812 bool 813 814config GENERIC_SCHED_CLOCK 815 bool 816 817menu "Scheduler features" 818 819config UCLAMP_TASK 820 bool "Enable utilization clamping for RT/FAIR tasks" 821 depends on CPU_FREQ_GOV_SCHEDUTIL 822 help 823 This feature enables the scheduler to track the clamped utilization 824 of each CPU based on RUNNABLE tasks scheduled on that CPU. 825 826 With this option, the user can specify the min and max CPU 827 utilization allowed for RUNNABLE tasks. The max utilization defines 828 the maximum frequency a task should use while the min utilization 829 defines the minimum frequency it should use. 830 831 Both min and max utilization clamp values are hints to the scheduler, 832 aiming at improving its frequency selection policy, but they do not 833 enforce or grant any specific bandwidth for tasks. 834 835 If in doubt, say N. 836 837config UCLAMP_BUCKETS_COUNT 838 int "Number of supported utilization clamp buckets" 839 range 5 20 840 default 5 841 depends on UCLAMP_TASK 842 help 843 Defines the number of clamp buckets to use. The range of each bucket 844 will be SCHED_CAPACITY_SCALE/UCLAMP_BUCKETS_COUNT. The higher the 845 number of clamp buckets the finer their granularity and the higher 846 the precision of clamping aggregation and tracking at run-time. 847 848 For example, with the minimum configuration value we will have 5 849 clamp buckets tracking 20% utilization each. A 25% boosted tasks will 850 be refcounted in the [20..39]% bucket and will set the bucket clamp 851 effective value to 25%. 852 If a second 30% boosted task should be co-scheduled on the same CPU, 853 that task will be refcounted in the same bucket of the first task and 854 it will boost the bucket clamp effective value to 30%. 855 The clamp effective value of a bucket is reset to its nominal value 856 (20% in the example above) when there are no more tasks refcounted in 857 that bucket. 858 859 An additional boost/capping margin can be added to some tasks. In the 860 example above the 25% task will be boosted to 30% until it exits the 861 CPU. If that should be considered not acceptable on certain systems, 862 it's always possible to reduce the margin by increasing the number of 863 clamp buckets to trade off used memory for run-time tracking 864 precision. 865 866 If in doubt, use the default value. 867 868endmenu 869 870# 871# For architectures that want to enable the support for NUMA-affine scheduler 872# balancing logic: 873# 874config ARCH_SUPPORTS_NUMA_BALANCING 875 bool 876 877# 878# For architectures that prefer to flush all TLBs after a number of pages 879# are unmapped instead of sending one IPI per page to flush. The architecture 880# must provide guarantees on what happens if a clean TLB cache entry is 881# written after the unmap. Details are in mm/rmap.c near the check for 882# should_defer_flush. The architecture should also consider if the full flush 883# and the refill costs are offset by the savings of sending fewer IPIs. 884config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH 885 bool 886 887config CC_HAS_INT128 888 def_bool !$(cc-option,$(m64-flag) -D__SIZEOF_INT128__=0) && 64BIT 889 890config CC_IMPLICIT_FALLTHROUGH 891 string 892 default "-Wimplicit-fallthrough=5" if CC_IS_GCC && $(cc-option,-Wimplicit-fallthrough=5) 893 default "-Wimplicit-fallthrough" if CC_IS_CLANG && $(cc-option,-Wunreachable-code-fallthrough) 894 895# Currently, disable gcc-11,12 array-bounds globally. 896# We may want to target only particular configurations some day. 897config GCC11_NO_ARRAY_BOUNDS 898 def_bool y 899 900config GCC12_NO_ARRAY_BOUNDS 901 def_bool y 902 903config CC_NO_ARRAY_BOUNDS 904 bool 905 default y if CC_IS_GCC && GCC_VERSION >= 110000 && GCC_VERSION < 120000 && GCC11_NO_ARRAY_BOUNDS 906 default y if CC_IS_GCC && GCC_VERSION >= 120000 && GCC_VERSION < 130000 && GCC12_NO_ARRAY_BOUNDS 907 908# 909# For architectures that know their GCC __int128 support is sound 910# 911config ARCH_SUPPORTS_INT128 912 bool 913 914# For architectures that (ab)use NUMA to represent different memory regions 915# all cpu-local but of different latencies, such as SuperH. 916# 917config ARCH_WANT_NUMA_VARIABLE_LOCALITY 918 bool 919 920config NUMA_BALANCING 921 bool "Memory placement aware NUMA scheduler" 922 depends on ARCH_SUPPORTS_NUMA_BALANCING 923 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY 924 depends on SMP && NUMA && MIGRATION && !PREEMPT_RT 925 help 926 This option adds support for automatic NUMA aware memory/task placement. 927 The mechanism is quite primitive and is based on migrating memory when 928 it has references to the node the task is running on. 929 930 This system will be inactive on UMA systems. 931 932config NUMA_BALANCING_DEFAULT_ENABLED 933 bool "Automatically enable NUMA aware memory/task placement" 934 default y 935 depends on NUMA_BALANCING 936 help 937 If set, automatic NUMA balancing will be enabled if running on a NUMA 938 machine. 939 940menuconfig CGROUPS 941 bool "Control Group support" 942 select KERNFS 943 help 944 This option adds support for grouping sets of processes together, for 945 use with process control subsystems such as Cpusets, CFS, memory 946 controls or device isolation. 947 See 948 - Documentation/scheduler/sched-design-CFS.rst (CFS) 949 - Documentation/admin-guide/cgroup-v1/ (features for grouping, isolation 950 and resource control) 951 952 Say N if unsure. 953 954if CGROUPS 955 956config PAGE_COUNTER 957 bool 958 959config CGROUP_FAVOR_DYNMODS 960 bool "Favor dynamic modification latency reduction by default" 961 help 962 This option enables the "favordynmods" mount option by default 963 which reduces the latencies of dynamic cgroup modifications such 964 as task migrations and controller on/offs at the cost of making 965 hot path operations such as forks and exits more expensive. 966 967 Say N if unsure. 968 969config MEMCG 970 bool "Memory controller" 971 select PAGE_COUNTER 972 select EVENTFD 973 help 974 Provides control over the memory footprint of tasks in a cgroup. 975 976config MEMCG_KMEM 977 bool 978 depends on MEMCG && !SLOB 979 default y 980 981config BLK_CGROUP 982 bool "IO controller" 983 depends on BLOCK 984 default n 985 help 986 Generic block IO controller cgroup interface. This is the common 987 cgroup interface which should be used by various IO controlling 988 policies. 989 990 Currently, CFQ IO scheduler uses it to recognize task groups and 991 control disk bandwidth allocation (proportional time slice allocation) 992 to such task groups. It is also used by bio throttling logic in 993 block layer to implement upper limit in IO rates on a device. 994 995 This option only enables generic Block IO controller infrastructure. 996 One needs to also enable actual IO controlling logic/policy. For 997 enabling proportional weight division of disk bandwidth in CFQ, set 998 CONFIG_BFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 999 CONFIG_BLK_DEV_THROTTLING=y. 1000 1001 See Documentation/admin-guide/cgroup-v1/blkio-controller.rst for more information. 1002 1003config CGROUP_WRITEBACK 1004 bool 1005 depends on MEMCG && BLK_CGROUP 1006 default y 1007 1008menuconfig CGROUP_SCHED 1009 bool "CPU controller" 1010 default n 1011 help 1012 This feature lets CPU scheduler recognize task groups and control CPU 1013 bandwidth allocation to such task groups. It uses cgroups to group 1014 tasks. 1015 1016if CGROUP_SCHED 1017config FAIR_GROUP_SCHED 1018 bool "Group scheduling for SCHED_OTHER" 1019 depends on CGROUP_SCHED 1020 default CGROUP_SCHED 1021 1022config CFS_BANDWIDTH 1023 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 1024 depends on FAIR_GROUP_SCHED 1025 default n 1026 help 1027 This option allows users to define CPU bandwidth rates (limits) for 1028 tasks running within the fair group scheduler. Groups with no limit 1029 set are considered to be unconstrained and will run with no 1030 restriction. 1031 See Documentation/scheduler/sched-bwc.rst for more information. 1032 1033config RT_GROUP_SCHED 1034 bool "Group scheduling for SCHED_RR/FIFO" 1035 depends on CGROUP_SCHED 1036 default n 1037 help 1038 This feature lets you explicitly allocate real CPU bandwidth 1039 to task groups. If enabled, it will also make it impossible to 1040 schedule realtime tasks for non-root users until you allocate 1041 realtime bandwidth for them. 1042 See Documentation/scheduler/sched-rt-group.rst for more information. 1043 1044endif #CGROUP_SCHED 1045 1046config UCLAMP_TASK_GROUP 1047 bool "Utilization clamping per group of tasks" 1048 depends on CGROUP_SCHED 1049 depends on UCLAMP_TASK 1050 default n 1051 help 1052 This feature enables the scheduler to track the clamped utilization 1053 of each CPU based on RUNNABLE tasks currently scheduled on that CPU. 1054 1055 When this option is enabled, the user can specify a min and max 1056 CPU bandwidth which is allowed for each single task in a group. 1057 The max bandwidth allows to clamp the maximum frequency a task 1058 can use, while the min bandwidth allows to define a minimum 1059 frequency a task will always use. 1060 1061 When task group based utilization clamping is enabled, an eventually 1062 specified task-specific clamp value is constrained by the cgroup 1063 specified clamp value. Both minimum and maximum task clamping cannot 1064 be bigger than the corresponding clamping defined at task group level. 1065 1066 If in doubt, say N. 1067 1068config CGROUP_PIDS 1069 bool "PIDs controller" 1070 help 1071 Provides enforcement of process number limits in the scope of a 1072 cgroup. Any attempt to fork more processes than is allowed in the 1073 cgroup will fail. PIDs are fundamentally a global resource because it 1074 is fairly trivial to reach PID exhaustion before you reach even a 1075 conservative kmemcg limit. As a result, it is possible to grind a 1076 system to halt without being limited by other cgroup policies. The 1077 PIDs controller is designed to stop this from happening. 1078 1079 It should be noted that organisational operations (such as attaching 1080 to a cgroup hierarchy) will *not* be blocked by the PIDs controller, 1081 since the PIDs limit only affects a process's ability to fork, not to 1082 attach to a cgroup. 1083 1084config CGROUP_RDMA 1085 bool "RDMA controller" 1086 help 1087 Provides enforcement of RDMA resources defined by IB stack. 1088 It is fairly easy for consumers to exhaust RDMA resources, which 1089 can result into resource unavailability to other consumers. 1090 RDMA controller is designed to stop this from happening. 1091 Attaching processes with active RDMA resources to the cgroup 1092 hierarchy is allowed even if can cross the hierarchy's limit. 1093 1094config CGROUP_FREEZER 1095 bool "Freezer controller" 1096 help 1097 Provides a way to freeze and unfreeze all tasks in a 1098 cgroup. 1099 1100 This option affects the ORIGINAL cgroup interface. The cgroup2 memory 1101 controller includes important in-kernel memory consumers per default. 1102 1103 If you're using cgroup2, say N. 1104 1105config CGROUP_HUGETLB 1106 bool "HugeTLB controller" 1107 depends on HUGETLB_PAGE 1108 select PAGE_COUNTER 1109 default n 1110 help 1111 Provides a cgroup controller for HugeTLB pages. 1112 When you enable this, you can put a per cgroup limit on HugeTLB usage. 1113 The limit is enforced during page fault. Since HugeTLB doesn't 1114 support page reclaim, enforcing the limit at page fault time implies 1115 that, the application will get SIGBUS signal if it tries to access 1116 HugeTLB pages beyond its limit. This requires the application to know 1117 beforehand how much HugeTLB pages it would require for its use. The 1118 control group is tracked in the third page lru pointer. This means 1119 that we cannot use the controller with huge page less than 3 pages. 1120 1121config CPUSETS 1122 bool "Cpuset controller" 1123 depends on SMP 1124 help 1125 This option will let you create and manage CPUSETs which 1126 allow dynamically partitioning a system into sets of CPUs and 1127 Memory Nodes and assigning tasks to run only within those sets. 1128 This is primarily useful on large SMP or NUMA systems. 1129 1130 Say N if unsure. 1131 1132config PROC_PID_CPUSET 1133 bool "Include legacy /proc/<pid>/cpuset file" 1134 depends on CPUSETS 1135 default y 1136 1137config CGROUP_DEVICE 1138 bool "Device controller" 1139 help 1140 Provides a cgroup controller implementing whitelists for 1141 devices which a process in the cgroup can mknod or open. 1142 1143config CGROUP_CPUACCT 1144 bool "Simple CPU accounting controller" 1145 help 1146 Provides a simple controller for monitoring the 1147 total CPU consumed by the tasks in a cgroup. 1148 1149config CGROUP_PERF 1150 bool "Perf controller" 1151 depends on PERF_EVENTS 1152 help 1153 This option extends the perf per-cpu mode to restrict monitoring 1154 to threads which belong to the cgroup specified and run on the 1155 designated cpu. Or this can be used to have cgroup ID in samples 1156 so that it can monitor performance events among cgroups. 1157 1158 Say N if unsure. 1159 1160config CGROUP_BPF 1161 bool "Support for eBPF programs attached to cgroups" 1162 depends on BPF_SYSCALL 1163 select SOCK_CGROUP_DATA 1164 help 1165 Allow attaching eBPF programs to a cgroup using the bpf(2) 1166 syscall command BPF_PROG_ATTACH. 1167 1168 In which context these programs are accessed depends on the type 1169 of attachment. For instance, programs that are attached using 1170 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of 1171 inet sockets. 1172 1173config CGROUP_MISC 1174 bool "Misc resource controller" 1175 default n 1176 help 1177 Provides a controller for miscellaneous resources on a host. 1178 1179 Miscellaneous scalar resources are the resources on the host system 1180 which cannot be abstracted like the other cgroups. This controller 1181 tracks and limits the miscellaneous resources used by a process 1182 attached to a cgroup hierarchy. 1183 1184 For more information, please check misc cgroup section in 1185 /Documentation/admin-guide/cgroup-v2.rst. 1186 1187config CGROUP_DEBUG 1188 bool "Debug controller" 1189 default n 1190 depends on DEBUG_KERNEL 1191 help 1192 This option enables a simple controller that exports 1193 debugging information about the cgroups framework. This 1194 controller is for control cgroup debugging only. Its 1195 interfaces are not stable. 1196 1197 Say N. 1198 1199config SOCK_CGROUP_DATA 1200 bool 1201 default n 1202 1203endif # CGROUPS 1204 1205menuconfig NAMESPACES 1206 bool "Namespaces support" if EXPERT 1207 depends on MULTIUSER 1208 default !EXPERT 1209 help 1210 Provides the way to make tasks work with different objects using 1211 the same id. For example same IPC id may refer to different objects 1212 or same user id or pid may refer to different tasks when used in 1213 different namespaces. 1214 1215if NAMESPACES 1216 1217config UTS_NS 1218 bool "UTS namespace" 1219 default y 1220 help 1221 In this namespace tasks see different info provided with the 1222 uname() system call 1223 1224config TIME_NS 1225 bool "TIME namespace" 1226 depends on GENERIC_VDSO_TIME_NS 1227 default y 1228 help 1229 In this namespace boottime and monotonic clocks can be set. 1230 The time will keep going with the same pace. 1231 1232config IPC_NS 1233 bool "IPC namespace" 1234 depends on (SYSVIPC || POSIX_MQUEUE) 1235 default y 1236 help 1237 In this namespace tasks work with IPC ids which correspond to 1238 different IPC objects in different namespaces. 1239 1240config USER_NS 1241 bool "User namespace" 1242 default n 1243 help 1244 This allows containers, i.e. vservers, to use user namespaces 1245 to provide different user info for different servers. 1246 1247 When user namespaces are enabled in the kernel it is 1248 recommended that the MEMCG option also be enabled and that 1249 user-space use the memory control groups to limit the amount 1250 of memory a memory unprivileged users can use. 1251 1252 If unsure, say N. 1253 1254config PID_NS 1255 bool "PID Namespaces" 1256 default y 1257 help 1258 Support process id namespaces. This allows having multiple 1259 processes with the same pid as long as they are in different 1260 pid namespaces. This is a building block of containers. 1261 1262config NET_NS 1263 bool "Network namespace" 1264 depends on NET 1265 default y 1266 help 1267 Allow user space to create what appear to be multiple instances 1268 of the network stack. 1269 1270endif # NAMESPACES 1271 1272config CHECKPOINT_RESTORE 1273 bool "Checkpoint/restore support" 1274 depends on PROC_FS 1275 select PROC_CHILDREN 1276 select KCMP 1277 default n 1278 help 1279 Enables additional kernel features in a sake of checkpoint/restore. 1280 In particular it adds auxiliary prctl codes to setup process text, 1281 data and heap segment sizes, and a few additional /proc filesystem 1282 entries. 1283 1284 If unsure, say N here. 1285 1286config SCHED_AUTOGROUP 1287 bool "Automatic process group scheduling" 1288 select CGROUPS 1289 select CGROUP_SCHED 1290 select FAIR_GROUP_SCHED 1291 help 1292 This option optimizes the scheduler for common desktop workloads by 1293 automatically creating and populating task groups. This separation 1294 of workloads isolates aggressive CPU burners (like build jobs) from 1295 desktop applications. Task group autogeneration is currently based 1296 upon task session. 1297 1298config SYSFS_DEPRECATED 1299 bool "Enable deprecated sysfs features to support old userspace tools" 1300 depends on SYSFS 1301 default n 1302 help 1303 This option adds code that switches the layout of the "block" class 1304 devices, to not show up in /sys/class/block/, but only in 1305 /sys/block/. 1306 1307 This switch is only active when the sysfs.deprecated=1 boot option is 1308 passed or the SYSFS_DEPRECATED_V2 option is set. 1309 1310 This option allows new kernels to run on old distributions and tools, 1311 which might get confused by /sys/class/block/. Since 2007/2008 all 1312 major distributions and tools handle this just fine. 1313 1314 Recent distributions and userspace tools after 2009/2010 depend on 1315 the existence of /sys/class/block/, and will not work with this 1316 option enabled. 1317 1318 Only if you are using a new kernel on an old distribution, you might 1319 need to say Y here. 1320 1321config SYSFS_DEPRECATED_V2 1322 bool "Enable deprecated sysfs features by default" 1323 default n 1324 depends on SYSFS 1325 depends on SYSFS_DEPRECATED 1326 help 1327 Enable deprecated sysfs by default. 1328 1329 See the CONFIG_SYSFS_DEPRECATED option for more details about this 1330 option. 1331 1332 Only if you are using a new kernel on an old distribution, you might 1333 need to say Y here. Even then, odds are you would not need it 1334 enabled, you can always pass the boot option if absolutely necessary. 1335 1336config RELAY 1337 bool "Kernel->user space relay support (formerly relayfs)" 1338 select IRQ_WORK 1339 help 1340 This option enables support for relay interface support in 1341 certain file systems (such as debugfs). 1342 It is designed to provide an efficient mechanism for tools and 1343 facilities to relay large amounts of data from kernel space to 1344 user space. 1345 1346 If unsure, say N. 1347 1348config BLK_DEV_INITRD 1349 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1350 help 1351 The initial RAM filesystem is a ramfs which is loaded by the 1352 boot loader (loadlin or lilo) and that is mounted as root 1353 before the normal boot procedure. It is typically used to 1354 load modules needed to mount the "real" root file system, 1355 etc. See <file:Documentation/admin-guide/initrd.rst> for details. 1356 1357 If RAM disk support (BLK_DEV_RAM) is also included, this 1358 also enables initial RAM disk (initrd) support and adds 1359 15 Kbytes (more on some other architectures) to the kernel size. 1360 1361 If unsure say Y. 1362 1363if BLK_DEV_INITRD 1364 1365source "usr/Kconfig" 1366 1367endif 1368 1369config BOOT_CONFIG 1370 bool "Boot config support" 1371 select BLK_DEV_INITRD if !BOOT_CONFIG_EMBED 1372 help 1373 Extra boot config allows system admin to pass a config file as 1374 complemental extension of kernel cmdline when booting. 1375 The boot config file must be attached at the end of initramfs 1376 with checksum, size and magic word. 1377 See <file:Documentation/admin-guide/bootconfig.rst> for details. 1378 1379 If unsure, say Y. 1380 1381config BOOT_CONFIG_EMBED 1382 bool "Embed bootconfig file in the kernel" 1383 depends on BOOT_CONFIG 1384 help 1385 Embed a bootconfig file given by BOOT_CONFIG_EMBED_FILE in the 1386 kernel. Usually, the bootconfig file is loaded with the initrd 1387 image. But if the system doesn't support initrd, this option will 1388 help you by embedding a bootconfig file while building the kernel. 1389 1390 If unsure, say N. 1391 1392config BOOT_CONFIG_EMBED_FILE 1393 string "Embedded bootconfig file path" 1394 depends on BOOT_CONFIG_EMBED 1395 help 1396 Specify a bootconfig file which will be embedded to the kernel. 1397 This bootconfig will be used if there is no initrd or no other 1398 bootconfig in the initrd. 1399 1400config INITRAMFS_PRESERVE_MTIME 1401 bool "Preserve cpio archive mtimes in initramfs" 1402 default y 1403 help 1404 Each entry in an initramfs cpio archive carries an mtime value. When 1405 enabled, extracted cpio items take this mtime, with directory mtime 1406 setting deferred until after creation of any child entries. 1407 1408 If unsure, say Y. 1409 1410choice 1411 prompt "Compiler optimization level" 1412 default CC_OPTIMIZE_FOR_PERFORMANCE 1413 1414config CC_OPTIMIZE_FOR_PERFORMANCE 1415 bool "Optimize for performance (-O2)" 1416 help 1417 This is the default optimization level for the kernel, building 1418 with the "-O2" compiler flag for best performance and most 1419 helpful compile-time warnings. 1420 1421config CC_OPTIMIZE_FOR_SIZE 1422 bool "Optimize for size (-Os)" 1423 help 1424 Choosing this option will pass "-Os" to your compiler resulting 1425 in a smaller kernel. 1426 1427endchoice 1428 1429config HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1430 bool 1431 help 1432 This requires that the arch annotates or otherwise protects 1433 its external entry points from being discarded. Linker scripts 1434 must also merge .text.*, .data.*, and .bss.* correctly into 1435 output sections. Care must be taken not to pull in unrelated 1436 sections (e.g., '.text.init'). Typically '.' in section names 1437 is used to distinguish them from label names / C identifiers. 1438 1439config LD_DEAD_CODE_DATA_ELIMINATION 1440 bool "Dead code and data elimination (EXPERIMENTAL)" 1441 depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1442 depends on EXPERT 1443 depends on $(cc-option,-ffunction-sections -fdata-sections) 1444 depends on $(ld-option,--gc-sections) 1445 help 1446 Enable this if you want to do dead code and data elimination with 1447 the linker by compiling with -ffunction-sections -fdata-sections, 1448 and linking with --gc-sections. 1449 1450 This can reduce on disk and in-memory size of the kernel 1451 code and static data, particularly for small configs and 1452 on small systems. This has the possibility of introducing 1453 silently broken kernel if the required annotations are not 1454 present. This option is not well tested yet, so use at your 1455 own risk. 1456 1457config LD_ORPHAN_WARN 1458 def_bool y 1459 depends on ARCH_WANT_LD_ORPHAN_WARN 1460 depends on $(ld-option,--orphan-handling=warn) 1461 1462config SYSCTL 1463 bool 1464 1465config HAVE_UID16 1466 bool 1467 1468config SYSCTL_EXCEPTION_TRACE 1469 bool 1470 help 1471 Enable support for /proc/sys/debug/exception-trace. 1472 1473config SYSCTL_ARCH_UNALIGN_NO_WARN 1474 bool 1475 help 1476 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap 1477 Allows arch to define/use @no_unaligned_warning to possibly warn 1478 about unaligned access emulation going on under the hood. 1479 1480config SYSCTL_ARCH_UNALIGN_ALLOW 1481 bool 1482 help 1483 Enable support for /proc/sys/kernel/unaligned-trap 1484 Allows arches to define/use @unaligned_enabled to runtime toggle 1485 the unaligned access emulation. 1486 see arch/parisc/kernel/unaligned.c for reference 1487 1488config HAVE_PCSPKR_PLATFORM 1489 bool 1490 1491# interpreter that classic socket filters depend on 1492config BPF 1493 bool 1494 select CRYPTO_LIB_SHA1 1495 1496menuconfig EXPERT 1497 bool "Configure standard kernel features (expert users)" 1498 # Unhide debug options, to make the on-by-default options visible 1499 select DEBUG_KERNEL 1500 help 1501 This option allows certain base kernel options and settings 1502 to be disabled or tweaked. This is for specialized 1503 environments which can tolerate a "non-standard" kernel. 1504 Only use this if you really know what you are doing. 1505 1506config UID16 1507 bool "Enable 16-bit UID system calls" if EXPERT 1508 depends on HAVE_UID16 && MULTIUSER 1509 default y 1510 help 1511 This enables the legacy 16-bit UID syscall wrappers. 1512 1513config MULTIUSER 1514 bool "Multiple users, groups and capabilities support" if EXPERT 1515 default y 1516 help 1517 This option enables support for non-root users, groups and 1518 capabilities. 1519 1520 If you say N here, all processes will run with UID 0, GID 0, and all 1521 possible capabilities. Saying N here also compiles out support for 1522 system calls related to UIDs, GIDs, and capabilities, such as setuid, 1523 setgid, and capset. 1524 1525 If unsure, say Y here. 1526 1527config SGETMASK_SYSCALL 1528 bool "sgetmask/ssetmask syscalls support" if EXPERT 1529 def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH 1530 help 1531 sys_sgetmask and sys_ssetmask are obsolete system calls 1532 no longer supported in libc but still enabled by default in some 1533 architectures. 1534 1535 If unsure, leave the default option here. 1536 1537config SYSFS_SYSCALL 1538 bool "Sysfs syscall support" if EXPERT 1539 default y 1540 help 1541 sys_sysfs is an obsolete system call no longer supported in libc. 1542 Note that disabling this option is more secure but might break 1543 compatibility with some systems. 1544 1545 If unsure say Y here. 1546 1547config FHANDLE 1548 bool "open by fhandle syscalls" if EXPERT 1549 select EXPORTFS 1550 default y 1551 help 1552 If you say Y here, a user level program will be able to map 1553 file names to handle and then later use the handle for 1554 different file system operations. This is useful in implementing 1555 userspace file servers, which now track files using handles instead 1556 of names. The handle would remain the same even if file names 1557 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 1558 syscalls. 1559 1560config POSIX_TIMERS 1561 bool "Posix Clocks & timers" if EXPERT 1562 default y 1563 help 1564 This includes native support for POSIX timers to the kernel. 1565 Some embedded systems have no use for them and therefore they 1566 can be configured out to reduce the size of the kernel image. 1567 1568 When this option is disabled, the following syscalls won't be 1569 available: timer_create, timer_gettime: timer_getoverrun, 1570 timer_settime, timer_delete, clock_adjtime, getitimer, 1571 setitimer, alarm. Furthermore, the clock_settime, clock_gettime, 1572 clock_getres and clock_nanosleep syscalls will be limited to 1573 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only. 1574 1575 If unsure say y. 1576 1577config PRINTK 1578 default y 1579 bool "Enable support for printk" if EXPERT 1580 select IRQ_WORK 1581 help 1582 This option enables normal printk support. Removing it 1583 eliminates most of the message strings from the kernel image 1584 and makes the kernel more or less silent. As this makes it 1585 very difficult to diagnose system problems, saying N here is 1586 strongly discouraged. 1587 1588config BUG 1589 bool "BUG() support" if EXPERT 1590 default y 1591 help 1592 Disabling this option eliminates support for BUG and WARN, reducing 1593 the size of your kernel image and potentially quietly ignoring 1594 numerous fatal conditions. You should only consider disabling this 1595 option for embedded systems with no facilities for reporting errors. 1596 Just say Y. 1597 1598config ELF_CORE 1599 depends on COREDUMP 1600 default y 1601 bool "Enable ELF core dumps" if EXPERT 1602 help 1603 Enable support for generating core dumps. Disabling saves about 4k. 1604 1605 1606config PCSPKR_PLATFORM 1607 bool "Enable PC-Speaker support" if EXPERT 1608 depends on HAVE_PCSPKR_PLATFORM 1609 select I8253_LOCK 1610 default y 1611 help 1612 This option allows to disable the internal PC-Speaker 1613 support, saving some memory. 1614 1615config BASE_FULL 1616 default y 1617 bool "Enable full-sized data structures for core" if EXPERT 1618 help 1619 Disabling this option reduces the size of miscellaneous core 1620 kernel data structures. This saves memory on small machines, 1621 but may reduce performance. 1622 1623config FUTEX 1624 bool "Enable futex support" if EXPERT 1625 depends on !(SPARC32 && SMP) 1626 default y 1627 imply RT_MUTEXES 1628 help 1629 Disabling this option will cause the kernel to be built without 1630 support for "fast userspace mutexes". The resulting kernel may not 1631 run glibc-based applications correctly. 1632 1633config FUTEX_PI 1634 bool 1635 depends on FUTEX && RT_MUTEXES 1636 default y 1637 1638config EPOLL 1639 bool "Enable eventpoll support" if EXPERT 1640 default y 1641 help 1642 Disabling this option will cause the kernel to be built without 1643 support for epoll family of system calls. 1644 1645config SIGNALFD 1646 bool "Enable signalfd() system call" if EXPERT 1647 default y 1648 help 1649 Enable the signalfd() system call that allows to receive signals 1650 on a file descriptor. 1651 1652 If unsure, say Y. 1653 1654config TIMERFD 1655 bool "Enable timerfd() system call" if EXPERT 1656 default y 1657 help 1658 Enable the timerfd() system call that allows to receive timer 1659 events on a file descriptor. 1660 1661 If unsure, say Y. 1662 1663config EVENTFD 1664 bool "Enable eventfd() system call" if EXPERT 1665 default y 1666 help 1667 Enable the eventfd() system call that allows to receive both 1668 kernel notification (ie. KAIO) or userspace notifications. 1669 1670 If unsure, say Y. 1671 1672config SHMEM 1673 bool "Use full shmem filesystem" if EXPERT 1674 default y 1675 depends on MMU 1676 help 1677 The shmem is an internal filesystem used to manage shared memory. 1678 It is backed by swap and manages resource limits. It is also exported 1679 to userspace as tmpfs if TMPFS is enabled. Disabling this 1680 option replaces shmem and tmpfs with the much simpler ramfs code, 1681 which may be appropriate on small systems without swap. 1682 1683config AIO 1684 bool "Enable AIO support" if EXPERT 1685 default y 1686 help 1687 This option enables POSIX asynchronous I/O which may by used 1688 by some high performance threaded applications. Disabling 1689 this option saves about 7k. 1690 1691config IO_URING 1692 bool "Enable IO uring support" if EXPERT 1693 select IO_WQ 1694 default y 1695 help 1696 This option enables support for the io_uring interface, enabling 1697 applications to submit and complete IO through submission and 1698 completion rings that are shared between the kernel and application. 1699 1700config ADVISE_SYSCALLS 1701 bool "Enable madvise/fadvise syscalls" if EXPERT 1702 default y 1703 help 1704 This option enables the madvise and fadvise syscalls, used by 1705 applications to advise the kernel about their future memory or file 1706 usage, improving performance. If building an embedded system where no 1707 applications use these syscalls, you can disable this option to save 1708 space. 1709 1710config MEMBARRIER 1711 bool "Enable membarrier() system call" if EXPERT 1712 default y 1713 help 1714 Enable the membarrier() system call that allows issuing memory 1715 barriers across all running threads, which can be used to distribute 1716 the cost of user-space memory barriers asymmetrically by transforming 1717 pairs of memory barriers into pairs consisting of membarrier() and a 1718 compiler barrier. 1719 1720 If unsure, say Y. 1721 1722config KALLSYMS 1723 bool "Load all symbols for debugging/ksymoops" if EXPERT 1724 default y 1725 help 1726 Say Y here to let the kernel print out symbolic crash information and 1727 symbolic stack backtraces. This increases the size of the kernel 1728 somewhat, as all symbols have to be loaded into the kernel image. 1729 1730config KALLSYMS_ALL 1731 bool "Include all symbols in kallsyms" 1732 depends on DEBUG_KERNEL && KALLSYMS 1733 help 1734 Normally kallsyms only contains the symbols of functions for nicer 1735 OOPS messages and backtraces (i.e., symbols from the text and inittext 1736 sections). This is sufficient for most cases. And only if you want to 1737 enable kernel live patching, or other less common use cases (e.g., 1738 when a debugger is used) all symbols are required (i.e., names of 1739 variables from the data sections, etc). 1740 1741 This option makes sure that all symbols are loaded into the kernel 1742 image (i.e., symbols from all sections) in cost of increased kernel 1743 size (depending on the kernel configuration, it may be 300KiB or 1744 something like this). 1745 1746 Say N unless you really need all symbols, or kernel live patching. 1747 1748config KALLSYMS_ABSOLUTE_PERCPU 1749 bool 1750 depends on KALLSYMS 1751 default X86_64 && SMP 1752 1753config KALLSYMS_BASE_RELATIVE 1754 bool 1755 depends on KALLSYMS 1756 default !IA64 1757 help 1758 Instead of emitting them as absolute values in the native word size, 1759 emit the symbol references in the kallsyms table as 32-bit entries, 1760 each containing a relative value in the range [base, base + U32_MAX] 1761 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either 1762 an absolute value in the range [0, S32_MAX] or a relative value in the 1763 range [base, base + S32_MAX], where base is the lowest relative symbol 1764 address encountered in the image. 1765 1766 On 64-bit builds, this reduces the size of the address table by 50%, 1767 but more importantly, it results in entries whose values are build 1768 time constants, and no relocation pass is required at runtime to fix 1769 up the entries based on the runtime load address of the kernel. 1770 1771# end of the "standard kernel features (expert users)" menu 1772 1773# syscall, maps, verifier 1774 1775config ARCH_HAS_MEMBARRIER_CALLBACKS 1776 bool 1777 1778config ARCH_HAS_MEMBARRIER_SYNC_CORE 1779 bool 1780 1781config KCMP 1782 bool "Enable kcmp() system call" if EXPERT 1783 help 1784 Enable the kernel resource comparison system call. It provides 1785 user-space with the ability to compare two processes to see if they 1786 share a common resource, such as a file descriptor or even virtual 1787 memory space. 1788 1789 If unsure, say N. 1790 1791config RSEQ 1792 bool "Enable rseq() system call" if EXPERT 1793 default y 1794 depends on HAVE_RSEQ 1795 select MEMBARRIER 1796 help 1797 Enable the restartable sequences system call. It provides a 1798 user-space cache for the current CPU number value, which 1799 speeds up getting the current CPU number from user-space, 1800 as well as an ABI to speed up user-space operations on 1801 per-CPU data. 1802 1803 If unsure, say Y. 1804 1805config DEBUG_RSEQ 1806 default n 1807 bool "Enabled debugging of rseq() system call" if EXPERT 1808 depends on RSEQ && DEBUG_KERNEL 1809 help 1810 Enable extra debugging checks for the rseq system call. 1811 1812 If unsure, say N. 1813 1814config EMBEDDED 1815 bool "Embedded system" 1816 select EXPERT 1817 help 1818 This option should be enabled if compiling the kernel for 1819 an embedded system so certain expert options are available 1820 for configuration. 1821 1822config HAVE_PERF_EVENTS 1823 bool 1824 help 1825 See tools/perf/design.txt for details. 1826 1827config GUEST_PERF_EVENTS 1828 bool 1829 depends on HAVE_PERF_EVENTS 1830 1831config PERF_USE_VMALLOC 1832 bool 1833 help 1834 See tools/perf/design.txt for details 1835 1836config PC104 1837 bool "PC/104 support" if EXPERT 1838 help 1839 Expose PC/104 form factor device drivers and options available for 1840 selection and configuration. Enable this option if your target 1841 machine has a PC/104 bus. 1842 1843menu "Kernel Performance Events And Counters" 1844 1845config PERF_EVENTS 1846 bool "Kernel performance events and counters" 1847 default y if PROFILING 1848 depends on HAVE_PERF_EVENTS 1849 select IRQ_WORK 1850 select SRCU 1851 help 1852 Enable kernel support for various performance events provided 1853 by software and hardware. 1854 1855 Software events are supported either built-in or via the 1856 use of generic tracepoints. 1857 1858 Most modern CPUs support performance events via performance 1859 counter registers. These registers count the number of certain 1860 types of hw events: such as instructions executed, cachemisses 1861 suffered, or branches mis-predicted - without slowing down the 1862 kernel or applications. These registers can also trigger interrupts 1863 when a threshold number of events have passed - and can thus be 1864 used to profile the code that runs on that CPU. 1865 1866 The Linux Performance Event subsystem provides an abstraction of 1867 these software and hardware event capabilities, available via a 1868 system call and used by the "perf" utility in tools/perf/. It 1869 provides per task and per CPU counters, and it provides event 1870 capabilities on top of those. 1871 1872 Say Y if unsure. 1873 1874config DEBUG_PERF_USE_VMALLOC 1875 default n 1876 bool "Debug: use vmalloc to back perf mmap() buffers" 1877 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC 1878 select PERF_USE_VMALLOC 1879 help 1880 Use vmalloc memory to back perf mmap() buffers. 1881 1882 Mostly useful for debugging the vmalloc code on platforms 1883 that don't require it. 1884 1885 Say N if unsure. 1886 1887endmenu 1888 1889config SYSTEM_DATA_VERIFICATION 1890 def_bool n 1891 select SYSTEM_TRUSTED_KEYRING 1892 select KEYS 1893 select CRYPTO 1894 select CRYPTO_RSA 1895 select ASYMMETRIC_KEY_TYPE 1896 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE 1897 select ASN1 1898 select OID_REGISTRY 1899 select X509_CERTIFICATE_PARSER 1900 select PKCS7_MESSAGE_PARSER 1901 help 1902 Provide PKCS#7 message verification using the contents of the system 1903 trusted keyring to provide public keys. This then can be used for 1904 module verification, kexec image verification and firmware blob 1905 verification. 1906 1907config PROFILING 1908 bool "Profiling support" 1909 help 1910 Say Y here to enable the extended profiling support mechanisms used 1911 by profilers. 1912 1913config RUST 1914 bool "Rust support" 1915 depends on HAVE_RUST 1916 depends on RUST_IS_AVAILABLE 1917 depends on !MODVERSIONS 1918 depends on !GCC_PLUGINS 1919 depends on !RANDSTRUCT 1920 depends on !DEBUG_INFO_BTF 1921 select CONSTRUCTORS 1922 help 1923 Enables Rust support in the kernel. 1924 1925 This allows other Rust-related options, like drivers written in Rust, 1926 to be selected. 1927 1928 It is also required to be able to load external kernel modules 1929 written in Rust. 1930 1931 See Documentation/rust/ for more information. 1932 1933 If unsure, say N. 1934 1935config RUSTC_VERSION_TEXT 1936 string 1937 depends on RUST 1938 default $(shell,command -v $(RUSTC) >/dev/null 2>&1 && $(RUSTC) --version || echo n) 1939 1940config BINDGEN_VERSION_TEXT 1941 string 1942 depends on RUST 1943 default $(shell,command -v $(BINDGEN) >/dev/null 2>&1 && $(BINDGEN) --version || echo n) 1944 1945# 1946# Place an empty function call at each tracepoint site. Can be 1947# dynamically changed for a probe function. 1948# 1949config TRACEPOINTS 1950 bool 1951 1952endmenu # General setup 1953 1954source "arch/Kconfig" 1955 1956config RT_MUTEXES 1957 bool 1958 default y if PREEMPT_RT 1959 1960config BASE_SMALL 1961 int 1962 default 0 if BASE_FULL 1963 default 1 if !BASE_FULL 1964 1965config MODULE_SIG_FORMAT 1966 def_bool n 1967 select SYSTEM_DATA_VERIFICATION 1968 1969source "kernel/module/Kconfig" 1970 1971config INIT_ALL_POSSIBLE 1972 bool 1973 help 1974 Back when each arch used to define their own cpu_online_mask and 1975 cpu_possible_mask, some of them chose to initialize cpu_possible_mask 1976 with all 1s, and others with all 0s. When they were centralised, 1977 it was better to provide this option than to break all the archs 1978 and have several arch maintainers pursuing me down dark alleys. 1979 1980source "block/Kconfig" 1981 1982config PREEMPT_NOTIFIERS 1983 bool 1984 1985config PADATA 1986 depends on SMP 1987 bool 1988 1989config ASN1 1990 tristate 1991 help 1992 Build a simple ASN.1 grammar compiler that produces a bytecode output 1993 that can be interpreted by the ASN.1 stream decoder and used to 1994 inform it as to what tags are to be expected in a stream and what 1995 functions to call on what tags. 1996 1997source "kernel/Kconfig.locks" 1998 1999config ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE 2000 bool 2001 2002config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE 2003 bool 2004 2005# It may be useful for an architecture to override the definitions of the 2006# SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h> 2007# and the COMPAT_ variants in <linux/compat.h>, in particular to use a 2008# different calling convention for syscalls. They can also override the 2009# macros for not-implemented syscalls in kernel/sys_ni.c and 2010# kernel/time/posix-stubs.c. All these overrides need to be available in 2011# <asm/syscall_wrapper.h>. 2012config ARCH_HAS_SYSCALL_WRAPPER 2013 def_bool n 2014