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 CC_CAN_LINK 64 bool 65 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag)) if 64BIT 66 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag)) 67 68config CC_CAN_LINK_STATIC 69 bool 70 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m64-flag) -static) if 64BIT 71 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(USERCFLAGS) $(USERLDFLAGS) $(m32-flag) -static) 72 73config CC_HAS_ASM_GOTO 74 def_bool $(success,$(srctree)/scripts/gcc-goto.sh $(CC)) 75 76config CC_HAS_ASM_GOTO_OUTPUT 77 depends on CC_HAS_ASM_GOTO 78 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) 79 80config TOOLS_SUPPORT_RELR 81 def_bool $(success,env "CC=$(CC)" "LD=$(LD)" "NM=$(NM)" "OBJCOPY=$(OBJCOPY)" $(srctree)/scripts/tools-support-relr.sh) 82 83config CC_HAS_ASM_INLINE 84 def_bool $(success,echo 'void foo(void) { asm inline (""); }' | $(CC) -x c - -c -o /dev/null) 85 86config CC_HAS_NO_PROFILE_FN_ATTR 87 def_bool $(success,echo '__attribute__((no_profile_instrument_function)) int x();' | $(CC) -x c - -c -o /dev/null -Werror) 88 89config PAHOLE_VERSION 90 int 91 default $(shell,$(srctree)/scripts/pahole-version.sh $(PAHOLE)) 92 93config CONSTRUCTORS 94 bool 95 96config IRQ_WORK 97 bool 98 99config BUILDTIME_TABLE_SORT 100 bool 101 102config THREAD_INFO_IN_TASK 103 bool 104 help 105 Select this to move thread_info off the stack into task_struct. To 106 make this work, an arch will need to remove all thread_info fields 107 except flags and fix any runtime bugs. 108 109 One subtle change that will be needed is to use try_get_task_stack() 110 and put_task_stack() in save_thread_stack_tsk() and get_wchan(). 111 112menu "General setup" 113 114config BROKEN 115 bool 116 117config BROKEN_ON_SMP 118 bool 119 depends on BROKEN || !SMP 120 default y 121 122config INIT_ENV_ARG_LIMIT 123 int 124 default 32 if !UML 125 default 128 if UML 126 help 127 Maximum of each of the number of arguments and environment 128 variables passed to init from the kernel command line. 129 130config COMPILE_TEST 131 bool "Compile also drivers which will not load" 132 depends on HAS_IOMEM 133 help 134 Some drivers can be compiled on a different platform than they are 135 intended to be run on. Despite they cannot be loaded there (or even 136 when they load they cannot be used due to missing HW support), 137 developers still, opposing to distributors, might want to build such 138 drivers to compile-test them. 139 140 If you are a developer and want to build everything available, say Y 141 here. If you are a user/distributor, say N here to exclude useless 142 drivers to be distributed. 143 144config WERROR 145 bool "Compile the kernel with warnings as errors" 146 default COMPILE_TEST 147 help 148 A kernel build should not cause any compiler warnings, and this 149 enables the '-Werror' flag to enforce that rule by default. 150 151 However, if you have a new (or very old) compiler with odd and 152 unusual warnings, or you have some architecture with problems, 153 you may need to disable this config option in order to 154 successfully build the kernel. 155 156 If in doubt, say Y. 157 158config UAPI_HEADER_TEST 159 bool "Compile test UAPI headers" 160 depends on HEADERS_INSTALL && CC_CAN_LINK 161 help 162 Compile test headers exported to user-space to ensure they are 163 self-contained, i.e. compilable as standalone units. 164 165 If you are a developer or tester and want to ensure the exported 166 headers are self-contained, say Y here. Otherwise, choose N. 167 168config LOCALVERSION 169 string "Local version - append to kernel release" 170 help 171 Append an extra string to the end of your kernel version. 172 This will show up when you type uname, for example. 173 The string you set here will be appended after the contents of 174 any files with a filename matching localversion* in your 175 object and source tree, in that order. Your total string can 176 be a maximum of 64 characters. 177 178config LOCALVERSION_AUTO 179 bool "Automatically append version information to the version string" 180 default y 181 depends on !COMPILE_TEST 182 help 183 This will try to automatically determine if the current tree is a 184 release tree by looking for git tags that belong to the current 185 top of tree revision. 186 187 A string of the format -gxxxxxxxx will be added to the localversion 188 if a git-based tree is found. The string generated by this will be 189 appended after any matching localversion* files, and after the value 190 set in CONFIG_LOCALVERSION. 191 192 (The actual string used here is the first eight characters produced 193 by running the command: 194 195 $ git rev-parse --verify HEAD 196 197 which is done within the script "scripts/setlocalversion".) 198 199config BUILD_SALT 200 string "Build ID Salt" 201 default "" 202 help 203 The build ID is used to link binaries and their debug info. Setting 204 this option will use the value in the calculation of the build id. 205 This is mostly useful for distributions which want to ensure the 206 build is unique between builds. It's safe to leave the default. 207 208config HAVE_KERNEL_GZIP 209 bool 210 211config HAVE_KERNEL_BZIP2 212 bool 213 214config HAVE_KERNEL_LZMA 215 bool 216 217config HAVE_KERNEL_XZ 218 bool 219 220config HAVE_KERNEL_LZO 221 bool 222 223config HAVE_KERNEL_LZ4 224 bool 225 226config HAVE_KERNEL_ZSTD 227 bool 228 229config HAVE_KERNEL_UNCOMPRESSED 230 bool 231 232choice 233 prompt "Kernel compression mode" 234 default KERNEL_GZIP 235 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 236 help 237 The linux kernel is a kind of self-extracting executable. 238 Several compression algorithms are available, which differ 239 in efficiency, compression and decompression speed. 240 Compression speed is only relevant when building a kernel. 241 Decompression speed is relevant at each boot. 242 243 If you have any problems with bzip2 or lzma compressed 244 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 245 version of this functionality (bzip2 only), for 2.4, was 246 supplied by Christian Ludwig) 247 248 High compression options are mostly useful for users, who 249 are low on disk space (embedded systems), but for whom ram 250 size matters less. 251 252 If in doubt, select 'gzip' 253 254config KERNEL_GZIP 255 bool "Gzip" 256 depends on HAVE_KERNEL_GZIP 257 help 258 The old and tried gzip compression. It provides a good balance 259 between compression ratio and decompression speed. 260 261config KERNEL_BZIP2 262 bool "Bzip2" 263 depends on HAVE_KERNEL_BZIP2 264 help 265 Its compression ratio and speed is intermediate. 266 Decompression speed is slowest among the choices. The kernel 267 size is about 10% smaller with bzip2, in comparison to gzip. 268 Bzip2 uses a large amount of memory. For modern kernels you 269 will need at least 8MB RAM or more for booting. 270 271config KERNEL_LZMA 272 bool "LZMA" 273 depends on HAVE_KERNEL_LZMA 274 help 275 This compression algorithm's ratio is best. Decompression speed 276 is between gzip and bzip2. Compression is slowest. 277 The kernel size is about 33% smaller with LZMA in comparison to gzip. 278 279config KERNEL_XZ 280 bool "XZ" 281 depends on HAVE_KERNEL_XZ 282 help 283 XZ uses the LZMA2 algorithm and instruction set specific 284 BCJ filters which can improve compression ratio of executable 285 code. The size of the kernel is about 30% smaller with XZ in 286 comparison to gzip. On architectures for which there is a BCJ 287 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 288 will create a few percent smaller kernel than plain LZMA. 289 290 The speed is about the same as with LZMA: The decompression 291 speed of XZ is better than that of bzip2 but worse than gzip 292 and LZO. Compression is slow. 293 294config KERNEL_LZO 295 bool "LZO" 296 depends on HAVE_KERNEL_LZO 297 help 298 Its compression ratio is the poorest among the choices. The kernel 299 size is about 10% bigger than gzip; however its speed 300 (both compression and decompression) is the fastest. 301 302config KERNEL_LZ4 303 bool "LZ4" 304 depends on HAVE_KERNEL_LZ4 305 help 306 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding. 307 A preliminary version of LZ4 de/compression tool is available at 308 <https://code.google.com/p/lz4/>. 309 310 Its compression ratio is worse than LZO. The size of the kernel 311 is about 8% bigger than LZO. But the decompression speed is 312 faster than LZO. 313 314config KERNEL_ZSTD 315 bool "ZSTD" 316 depends on HAVE_KERNEL_ZSTD 317 help 318 ZSTD is a compression algorithm targeting intermediate compression 319 with fast decompression speed. It will compress better than GZIP and 320 decompress around the same speed as LZO, but slower than LZ4. You 321 will need at least 192 KB RAM or more for booting. The zstd command 322 line tool is required for compression. 323 324config KERNEL_UNCOMPRESSED 325 bool "None" 326 depends on HAVE_KERNEL_UNCOMPRESSED 327 help 328 Produce uncompressed kernel image. This option is usually not what 329 you want. It is useful for debugging the kernel in slow simulation 330 environments, where decompressing and moving the kernel is awfully 331 slow. This option allows early boot code to skip the decompressor 332 and jump right at uncompressed kernel image. 333 334endchoice 335 336config DEFAULT_INIT 337 string "Default init path" 338 default "" 339 help 340 This option determines the default init for the system if no init= 341 option is passed on the kernel command line. If the requested path is 342 not present, we will still then move on to attempting further 343 locations (e.g. /sbin/init, etc). If this is empty, we will just use 344 the fallback list when init= is not passed. 345 346config DEFAULT_HOSTNAME 347 string "Default hostname" 348 default "(none)" 349 help 350 This option determines the default system hostname before userspace 351 calls sethostname(2). The kernel traditionally uses "(none)" here, 352 but you may wish to use a different default here to make a minimal 353 system more usable with less configuration. 354 355# 356# For some reason microblaze and nios2 hard code SWAP=n. Hopefully we can 357# add proper SWAP support to them, in which case this can be remove. 358# 359config ARCH_NO_SWAP 360 bool 361 362config SWAP 363 bool "Support for paging of anonymous memory (swap)" 364 depends on MMU && BLOCK && !ARCH_NO_SWAP 365 default y 366 help 367 This option allows you to choose whether you want to have support 368 for so called swap devices or swap files in your kernel that are 369 used to provide more virtual memory than the actual RAM present 370 in your computer. If unsure say Y. 371 372config SYSVIPC 373 bool "System V IPC" 374 help 375 Inter Process Communication is a suite of library functions and 376 system calls which let processes (running programs) synchronize and 377 exchange information. It is generally considered to be a good thing, 378 and some programs won't run unless you say Y here. In particular, if 379 you want to run the DOS emulator dosemu under Linux (read the 380 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 381 you'll need to say Y here. 382 383 You can find documentation about IPC with "info ipc" and also in 384 section 6.4 of the Linux Programmer's Guide, available from 385 <http://www.tldp.org/guides.html>. 386 387config SYSVIPC_SYSCTL 388 bool 389 depends on SYSVIPC 390 depends on SYSCTL 391 default y 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/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 if !PPC64 477 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64 478 479# Kind of a stub config for the pure tick based cputime accounting 480config TICK_CPU_ACCOUNTING 481 bool "Simple tick based cputime accounting" 482 depends on !S390 && !NO_HZ_FULL 483 help 484 This is the basic tick based cputime accounting that maintains 485 statistics about user, system and idle time spent on per jiffies 486 granularity. 487 488 If unsure, say Y. 489 490config VIRT_CPU_ACCOUNTING_NATIVE 491 bool "Deterministic task and CPU time accounting" 492 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL 493 select VIRT_CPU_ACCOUNTING 494 help 495 Select this option to enable more accurate task and CPU time 496 accounting. This is done by reading a CPU counter on each 497 kernel entry and exit and on transitions within the kernel 498 between system, softirq and hardirq state, so there is a 499 small performance impact. In the case of s390 or IBM POWER > 5, 500 this also enables accounting of stolen time on logically-partitioned 501 systems. 502 503config VIRT_CPU_ACCOUNTING_GEN 504 bool "Full dynticks CPU time accounting" 505 depends on HAVE_CONTEXT_TRACKING 506 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN 507 depends on GENERIC_CLOCKEVENTS 508 select VIRT_CPU_ACCOUNTING 509 select CONTEXT_TRACKING 510 help 511 Select this option to enable task and CPU time accounting on full 512 dynticks systems. This accounting is implemented by watching every 513 kernel-user boundaries using the context tracking subsystem. 514 The accounting is thus performed at the expense of some significant 515 overhead. 516 517 For now this is only useful if you are working on the full 518 dynticks subsystem development. 519 520 If unsure, say N. 521 522endchoice 523 524config IRQ_TIME_ACCOUNTING 525 bool "Fine granularity task level IRQ time accounting" 526 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE 527 help 528 Select this option to enable fine granularity task irq time 529 accounting. This is done by reading a timestamp on each 530 transitions between softirq and hardirq state, so there can be a 531 small performance impact. 532 533 If in doubt, say N here. 534 535config HAVE_SCHED_AVG_IRQ 536 def_bool y 537 depends on IRQ_TIME_ACCOUNTING || PARAVIRT_TIME_ACCOUNTING 538 depends on SMP 539 540config SCHED_THERMAL_PRESSURE 541 bool 542 default y if ARM && ARM_CPU_TOPOLOGY 543 default y if ARM64 544 depends on SMP 545 depends on CPU_FREQ_THERMAL 546 help 547 Select this option to enable thermal pressure accounting in the 548 scheduler. Thermal pressure is the value conveyed to the scheduler 549 that reflects the reduction in CPU compute capacity resulted from 550 thermal throttling. Thermal throttling occurs when the performance of 551 a CPU is capped due to high operating temperatures. 552 553 If selected, the scheduler will be able to balance tasks accordingly, 554 i.e. put less load on throttled CPUs than on non/less throttled ones. 555 556 This requires the architecture to implement 557 arch_update_thermal_pressure() and arch_scale_thermal_pressure(). 558 559config BSD_PROCESS_ACCT 560 bool "BSD Process Accounting" 561 depends on MULTIUSER 562 help 563 If you say Y here, a user level program will be able to instruct the 564 kernel (via a special system call) to write process accounting 565 information to a file: whenever a process exits, information about 566 that process will be appended to the file by the kernel. The 567 information includes things such as creation time, owning user, 568 command name, memory usage, controlling terminal etc. (the complete 569 list is in the struct acct in <file:include/linux/acct.h>). It is 570 up to the user level program to do useful things with this 571 information. This is generally a good idea, so say Y. 572 573config BSD_PROCESS_ACCT_V3 574 bool "BSD Process Accounting version 3 file format" 575 depends on BSD_PROCESS_ACCT 576 default n 577 help 578 If you say Y here, the process accounting information is written 579 in a new file format that also logs the process IDs of each 580 process and its parent. Note that this file format is incompatible 581 with previous v0/v1/v2 file formats, so you will need updated tools 582 for processing it. A preliminary version of these tools is available 583 at <http://www.gnu.org/software/acct/>. 584 585config TASKSTATS 586 bool "Export task/process statistics through netlink" 587 depends on NET 588 depends on MULTIUSER 589 default n 590 help 591 Export selected statistics for tasks/processes through the 592 generic netlink interface. Unlike BSD process accounting, the 593 statistics are available during the lifetime of tasks/processes as 594 responses to commands. Like BSD accounting, they are sent to user 595 space on task exit. 596 597 Say N if unsure. 598 599config TASK_DELAY_ACCT 600 bool "Enable per-task delay accounting" 601 depends on TASKSTATS 602 select SCHED_INFO 603 help 604 Collect information on time spent by a task waiting for system 605 resources like cpu, synchronous block I/O completion and swapping 606 in pages. Such statistics can help in setting a task's priorities 607 relative to other tasks for cpu, io, rss limits etc. 608 609 Say N if unsure. 610 611config TASK_XACCT 612 bool "Enable extended accounting over taskstats" 613 depends on TASKSTATS 614 help 615 Collect extended task accounting data and send the data 616 to userland for processing over the taskstats interface. 617 618 Say N if unsure. 619 620config TASK_IO_ACCOUNTING 621 bool "Enable per-task storage I/O accounting" 622 depends on TASK_XACCT 623 help 624 Collect information on the number of bytes of storage I/O which this 625 task has caused. 626 627 Say N if unsure. 628 629config PSI 630 bool "Pressure stall information tracking" 631 help 632 Collect metrics that indicate how overcommitted the CPU, memory, 633 and IO capacity are in the system. 634 635 If you say Y here, the kernel will create /proc/pressure/ with the 636 pressure statistics files cpu, memory, and io. These will indicate 637 the share of walltime in which some or all tasks in the system are 638 delayed due to contention of the respective resource. 639 640 In kernels with cgroup support, cgroups (cgroup2 only) will 641 have cpu.pressure, memory.pressure, and io.pressure files, 642 which aggregate pressure stalls for the grouped tasks only. 643 644 For more details see Documentation/accounting/psi.rst. 645 646 Say N if unsure. 647 648config PSI_DEFAULT_DISABLED 649 bool "Require boot parameter to enable pressure stall information tracking" 650 default n 651 depends on PSI 652 help 653 If set, pressure stall information tracking will be disabled 654 per default but can be enabled through passing psi=1 on the 655 kernel commandline during boot. 656 657 This feature adds some code to the task wakeup and sleep 658 paths of the scheduler. The overhead is too low to affect 659 common scheduling-intense workloads in practice (such as 660 webservers, memcache), but it does show up in artificial 661 scheduler stress tests, such as hackbench. 662 663 If you are paranoid and not sure what the kernel will be 664 used for, say Y. 665 666 Say N if unsure. 667 668endmenu # "CPU/Task time and stats accounting" 669 670config CPU_ISOLATION 671 bool "CPU isolation" 672 depends on SMP || COMPILE_TEST 673 default y 674 help 675 Make sure that CPUs running critical tasks are not disturbed by 676 any source of "noise" such as unbound workqueues, timers, kthreads... 677 Unbound jobs get offloaded to housekeeping CPUs. This is driven by 678 the "isolcpus=" boot parameter. 679 680 Say Y if unsure. 681 682source "kernel/rcu/Kconfig" 683 684config BUILD_BIN2C 685 bool 686 default n 687 688config IKCONFIG 689 tristate "Kernel .config support" 690 help 691 This option enables the complete Linux kernel ".config" file 692 contents to be saved in the kernel. It provides documentation 693 of which kernel options are used in a running kernel or in an 694 on-disk kernel. This information can be extracted from the kernel 695 image file with the script scripts/extract-ikconfig and used as 696 input to rebuild the current kernel or to build another kernel. 697 It can also be extracted from a running kernel by reading 698 /proc/config.gz if enabled (below). 699 700config IKCONFIG_PROC 701 bool "Enable access to .config through /proc/config.gz" 702 depends on IKCONFIG && PROC_FS 703 help 704 This option enables access to the kernel configuration file 705 through /proc/config.gz. 706 707config IKHEADERS 708 tristate "Enable kernel headers through /sys/kernel/kheaders.tar.xz" 709 depends on SYSFS 710 help 711 This option enables access to the in-kernel headers that are generated during 712 the build process. These can be used to build eBPF tracing programs, 713 or similar programs. If you build the headers as a module, a module called 714 kheaders.ko is built which can be loaded on-demand to get access to headers. 715 716config LOG_BUF_SHIFT 717 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 718 range 12 25 if !H8300 719 range 12 19 if H8300 720 default 17 721 depends on PRINTK 722 help 723 Select the minimal kernel log buffer size as a power of 2. 724 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config 725 parameter, see below. Any higher size also might be forced 726 by "log_buf_len" boot parameter. 727 728 Examples: 729 17 => 128 KB 730 16 => 64 KB 731 15 => 32 KB 732 14 => 16 KB 733 13 => 8 KB 734 12 => 4 KB 735 736config LOG_CPU_MAX_BUF_SHIFT 737 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)" 738 depends on SMP 739 range 0 21 740 default 12 if !BASE_SMALL 741 default 0 if BASE_SMALL 742 depends on PRINTK 743 help 744 This option allows to increase the default ring buffer size 745 according to the number of CPUs. The value defines the contribution 746 of each CPU as a power of 2. The used space is typically only few 747 lines however it might be much more when problems are reported, 748 e.g. backtraces. 749 750 The increased size means that a new buffer has to be allocated and 751 the original static one is unused. It makes sense only on systems 752 with more CPUs. Therefore this value is used only when the sum of 753 contributions is greater than the half of the default kernel ring 754 buffer as defined by LOG_BUF_SHIFT. The default values are set 755 so that more than 16 CPUs are needed to trigger the allocation. 756 757 Also this option is ignored when "log_buf_len" kernel parameter is 758 used as it forces an exact (power of two) size of the ring buffer. 759 760 The number of possible CPUs is used for this computation ignoring 761 hotplugging making the computation optimal for the worst case 762 scenario while allowing a simple algorithm to be used from bootup. 763 764 Examples shift values and their meaning: 765 17 => 128 KB for each CPU 766 16 => 64 KB for each CPU 767 15 => 32 KB for each CPU 768 14 => 16 KB for each CPU 769 13 => 8 KB for each CPU 770 12 => 4 KB for each CPU 771 772config PRINTK_SAFE_LOG_BUF_SHIFT 773 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)" 774 range 10 21 775 default 13 776 depends on PRINTK 777 help 778 Select the size of an alternate printk per-CPU buffer where messages 779 printed from usafe contexts are temporary stored. One example would 780 be NMI messages, another one - printk recursion. The messages are 781 copied to the main log buffer in a safe context to avoid a deadlock. 782 The value defines the size as a power of 2. 783 784 Those messages are rare and limited. The largest one is when 785 a backtrace is printed. It usually fits into 4KB. Select 786 8KB if you want to be on the safe side. 787 788 Examples: 789 17 => 128 KB for each CPU 790 16 => 64 KB for each CPU 791 15 => 32 KB for each CPU 792 14 => 16 KB for each CPU 793 13 => 8 KB for each CPU 794 12 => 4 KB for each CPU 795 796config PRINTK_INDEX 797 bool "Printk indexing debugfs interface" 798 depends on PRINTK && DEBUG_FS 799 help 800 Add support for indexing of all printk formats known at compile time 801 at <debugfs>/printk/index/<module>. 802 803 This can be used as part of maintaining daemons which monitor 804 /dev/kmsg, as it permits auditing the printk formats present in a 805 kernel, allowing detection of cases where monitored printks are 806 changed or no longer present. 807 808 There is no additional runtime cost to printk with this enabled. 809 810# 811# Architectures with an unreliable sched_clock() should select this: 812# 813config HAVE_UNSTABLE_SCHED_CLOCK 814 bool 815 816config GENERIC_SCHED_CLOCK 817 bool 818 819menu "Scheduler features" 820 821config UCLAMP_TASK 822 bool "Enable utilization clamping for RT/FAIR tasks" 823 depends on CPU_FREQ_GOV_SCHEDUTIL 824 help 825 This feature enables the scheduler to track the clamped utilization 826 of each CPU based on RUNNABLE tasks scheduled on that CPU. 827 828 With this option, the user can specify the min and max CPU 829 utilization allowed for RUNNABLE tasks. The max utilization defines 830 the maximum frequency a task should use while the min utilization 831 defines the minimum frequency it should use. 832 833 Both min and max utilization clamp values are hints to the scheduler, 834 aiming at improving its frequency selection policy, but they do not 835 enforce or grant any specific bandwidth for tasks. 836 837 If in doubt, say N. 838 839config UCLAMP_BUCKETS_COUNT 840 int "Number of supported utilization clamp buckets" 841 range 5 20 842 default 5 843 depends on UCLAMP_TASK 844 help 845 Defines the number of clamp buckets to use. The range of each bucket 846 will be SCHED_CAPACITY_SCALE/UCLAMP_BUCKETS_COUNT. The higher the 847 number of clamp buckets the finer their granularity and the higher 848 the precision of clamping aggregation and tracking at run-time. 849 850 For example, with the minimum configuration value we will have 5 851 clamp buckets tracking 20% utilization each. A 25% boosted tasks will 852 be refcounted in the [20..39]% bucket and will set the bucket clamp 853 effective value to 25%. 854 If a second 30% boosted task should be co-scheduled on the same CPU, 855 that task will be refcounted in the same bucket of the first task and 856 it will boost the bucket clamp effective value to 30%. 857 The clamp effective value of a bucket is reset to its nominal value 858 (20% in the example above) when there are no more tasks refcounted in 859 that bucket. 860 861 An additional boost/capping margin can be added to some tasks. In the 862 example above the 25% task will be boosted to 30% until it exits the 863 CPU. If that should be considered not acceptable on certain systems, 864 it's always possible to reduce the margin by increasing the number of 865 clamp buckets to trade off used memory for run-time tracking 866 precision. 867 868 If in doubt, use the default value. 869 870endmenu 871 872# 873# For architectures that want to enable the support for NUMA-affine scheduler 874# balancing logic: 875# 876config ARCH_SUPPORTS_NUMA_BALANCING 877 bool 878 879# 880# For architectures that prefer to flush all TLBs after a number of pages 881# are unmapped instead of sending one IPI per page to flush. The architecture 882# must provide guarantees on what happens if a clean TLB cache entry is 883# written after the unmap. Details are in mm/rmap.c near the check for 884# should_defer_flush. The architecture should also consider if the full flush 885# and the refill costs are offset by the savings of sending fewer IPIs. 886config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH 887 bool 888 889config CC_HAS_INT128 890 def_bool !$(cc-option,$(m64-flag) -D__SIZEOF_INT128__=0) && 64BIT 891 892config CC_IMPLICIT_FALLTHROUGH 893 string 894 default "-Wimplicit-fallthrough=5" if CC_IS_GCC && $(cc-option,-Wimplicit-fallthrough=5) 895 default "-Wimplicit-fallthrough" if CC_IS_CLANG && $(cc-option,-Wunreachable-code-fallthrough) 896 897# 898# For architectures that know their GCC __int128 support is sound 899# 900config ARCH_SUPPORTS_INT128 901 bool 902 903# For architectures that (ab)use NUMA to represent different memory regions 904# all cpu-local but of different latencies, such as SuperH. 905# 906config ARCH_WANT_NUMA_VARIABLE_LOCALITY 907 bool 908 909config NUMA_BALANCING 910 bool "Memory placement aware NUMA scheduler" 911 depends on ARCH_SUPPORTS_NUMA_BALANCING 912 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY 913 depends on SMP && NUMA && MIGRATION && !PREEMPT_RT 914 help 915 This option adds support for automatic NUMA aware memory/task placement. 916 The mechanism is quite primitive and is based on migrating memory when 917 it has references to the node the task is running on. 918 919 This system will be inactive on UMA systems. 920 921config NUMA_BALANCING_DEFAULT_ENABLED 922 bool "Automatically enable NUMA aware memory/task placement" 923 default y 924 depends on NUMA_BALANCING 925 help 926 If set, automatic NUMA balancing will be enabled if running on a NUMA 927 machine. 928 929menuconfig CGROUPS 930 bool "Control Group support" 931 select KERNFS 932 help 933 This option adds support for grouping sets of processes together, for 934 use with process control subsystems such as Cpusets, CFS, memory 935 controls or device isolation. 936 See 937 - Documentation/scheduler/sched-design-CFS.rst (CFS) 938 - Documentation/admin-guide/cgroup-v1/ (features for grouping, isolation 939 and resource control) 940 941 Say N if unsure. 942 943if CGROUPS 944 945config PAGE_COUNTER 946 bool 947 948config MEMCG 949 bool "Memory controller" 950 select PAGE_COUNTER 951 select EVENTFD 952 help 953 Provides control over the memory footprint of tasks in a cgroup. 954 955config MEMCG_SWAP 956 bool 957 depends on MEMCG && SWAP 958 default y 959 960config MEMCG_KMEM 961 bool 962 depends on MEMCG && !SLOB 963 default y 964 965config BLK_CGROUP 966 bool "IO controller" 967 depends on BLOCK 968 default n 969 help 970 Generic block IO controller cgroup interface. This is the common 971 cgroup interface which should be used by various IO controlling 972 policies. 973 974 Currently, CFQ IO scheduler uses it to recognize task groups and 975 control disk bandwidth allocation (proportional time slice allocation) 976 to such task groups. It is also used by bio throttling logic in 977 block layer to implement upper limit in IO rates on a device. 978 979 This option only enables generic Block IO controller infrastructure. 980 One needs to also enable actual IO controlling logic/policy. For 981 enabling proportional weight division of disk bandwidth in CFQ, set 982 CONFIG_BFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 983 CONFIG_BLK_DEV_THROTTLING=y. 984 985 See Documentation/admin-guide/cgroup-v1/blkio-controller.rst for more information. 986 987config CGROUP_WRITEBACK 988 bool 989 depends on MEMCG && BLK_CGROUP 990 default y 991 992menuconfig CGROUP_SCHED 993 bool "CPU controller" 994 default n 995 help 996 This feature lets CPU scheduler recognize task groups and control CPU 997 bandwidth allocation to such task groups. It uses cgroups to group 998 tasks. 999 1000if CGROUP_SCHED 1001config FAIR_GROUP_SCHED 1002 bool "Group scheduling for SCHED_OTHER" 1003 depends on CGROUP_SCHED 1004 default CGROUP_SCHED 1005 1006config CFS_BANDWIDTH 1007 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 1008 depends on FAIR_GROUP_SCHED 1009 default n 1010 help 1011 This option allows users to define CPU bandwidth rates (limits) for 1012 tasks running within the fair group scheduler. Groups with no limit 1013 set are considered to be unconstrained and will run with no 1014 restriction. 1015 See Documentation/scheduler/sched-bwc.rst for more information. 1016 1017config RT_GROUP_SCHED 1018 bool "Group scheduling for SCHED_RR/FIFO" 1019 depends on CGROUP_SCHED 1020 default n 1021 help 1022 This feature lets you explicitly allocate real CPU bandwidth 1023 to task groups. If enabled, it will also make it impossible to 1024 schedule realtime tasks for non-root users until you allocate 1025 realtime bandwidth for them. 1026 See Documentation/scheduler/sched-rt-group.rst for more information. 1027 1028endif #CGROUP_SCHED 1029 1030config UCLAMP_TASK_GROUP 1031 bool "Utilization clamping per group of tasks" 1032 depends on CGROUP_SCHED 1033 depends on UCLAMP_TASK 1034 default n 1035 help 1036 This feature enables the scheduler to track the clamped utilization 1037 of each CPU based on RUNNABLE tasks currently scheduled on that CPU. 1038 1039 When this option is enabled, the user can specify a min and max 1040 CPU bandwidth which is allowed for each single task in a group. 1041 The max bandwidth allows to clamp the maximum frequency a task 1042 can use, while the min bandwidth allows to define a minimum 1043 frequency a task will always use. 1044 1045 When task group based utilization clamping is enabled, an eventually 1046 specified task-specific clamp value is constrained by the cgroup 1047 specified clamp value. Both minimum and maximum task clamping cannot 1048 be bigger than the corresponding clamping defined at task group level. 1049 1050 If in doubt, say N. 1051 1052config CGROUP_PIDS 1053 bool "PIDs controller" 1054 help 1055 Provides enforcement of process number limits in the scope of a 1056 cgroup. Any attempt to fork more processes than is allowed in the 1057 cgroup will fail. PIDs are fundamentally a global resource because it 1058 is fairly trivial to reach PID exhaustion before you reach even a 1059 conservative kmemcg limit. As a result, it is possible to grind a 1060 system to halt without being limited by other cgroup policies. The 1061 PIDs controller is designed to stop this from happening. 1062 1063 It should be noted that organisational operations (such as attaching 1064 to a cgroup hierarchy) will *not* be blocked by the PIDs controller, 1065 since the PIDs limit only affects a process's ability to fork, not to 1066 attach to a cgroup. 1067 1068config CGROUP_RDMA 1069 bool "RDMA controller" 1070 help 1071 Provides enforcement of RDMA resources defined by IB stack. 1072 It is fairly easy for consumers to exhaust RDMA resources, which 1073 can result into resource unavailability to other consumers. 1074 RDMA controller is designed to stop this from happening. 1075 Attaching processes with active RDMA resources to the cgroup 1076 hierarchy is allowed even if can cross the hierarchy's limit. 1077 1078config CGROUP_FREEZER 1079 bool "Freezer controller" 1080 help 1081 Provides a way to freeze and unfreeze all tasks in a 1082 cgroup. 1083 1084 This option affects the ORIGINAL cgroup interface. The cgroup2 memory 1085 controller includes important in-kernel memory consumers per default. 1086 1087 If you're using cgroup2, say N. 1088 1089config CGROUP_HUGETLB 1090 bool "HugeTLB controller" 1091 depends on HUGETLB_PAGE 1092 select PAGE_COUNTER 1093 default n 1094 help 1095 Provides a cgroup controller for HugeTLB pages. 1096 When you enable this, you can put a per cgroup limit on HugeTLB usage. 1097 The limit is enforced during page fault. Since HugeTLB doesn't 1098 support page reclaim, enforcing the limit at page fault time implies 1099 that, the application will get SIGBUS signal if it tries to access 1100 HugeTLB pages beyond its limit. This requires the application to know 1101 beforehand how much HugeTLB pages it would require for its use. The 1102 control group is tracked in the third page lru pointer. This means 1103 that we cannot use the controller with huge page less than 3 pages. 1104 1105config CPUSETS 1106 bool "Cpuset controller" 1107 depends on SMP 1108 help 1109 This option will let you create and manage CPUSETs which 1110 allow dynamically partitioning a system into sets of CPUs and 1111 Memory Nodes and assigning tasks to run only within those sets. 1112 This is primarily useful on large SMP or NUMA systems. 1113 1114 Say N if unsure. 1115 1116config PROC_PID_CPUSET 1117 bool "Include legacy /proc/<pid>/cpuset file" 1118 depends on CPUSETS 1119 default y 1120 1121config CGROUP_DEVICE 1122 bool "Device controller" 1123 help 1124 Provides a cgroup controller implementing whitelists for 1125 devices which a process in the cgroup can mknod or open. 1126 1127config CGROUP_CPUACCT 1128 bool "Simple CPU accounting controller" 1129 help 1130 Provides a simple controller for monitoring the 1131 total CPU consumed by the tasks in a cgroup. 1132 1133config CGROUP_PERF 1134 bool "Perf controller" 1135 depends on PERF_EVENTS 1136 help 1137 This option extends the perf per-cpu mode to restrict monitoring 1138 to threads which belong to the cgroup specified and run on the 1139 designated cpu. Or this can be used to have cgroup ID in samples 1140 so that it can monitor performance events among cgroups. 1141 1142 Say N if unsure. 1143 1144config CGROUP_BPF 1145 bool "Support for eBPF programs attached to cgroups" 1146 depends on BPF_SYSCALL 1147 select SOCK_CGROUP_DATA 1148 help 1149 Allow attaching eBPF programs to a cgroup using the bpf(2) 1150 syscall command BPF_PROG_ATTACH. 1151 1152 In which context these programs are accessed depends on the type 1153 of attachment. For instance, programs that are attached using 1154 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of 1155 inet sockets. 1156 1157config CGROUP_MISC 1158 bool "Misc resource controller" 1159 default n 1160 help 1161 Provides a controller for miscellaneous resources on a host. 1162 1163 Miscellaneous scalar resources are the resources on the host system 1164 which cannot be abstracted like the other cgroups. This controller 1165 tracks and limits the miscellaneous resources used by a process 1166 attached to a cgroup hierarchy. 1167 1168 For more information, please check misc cgroup section in 1169 /Documentation/admin-guide/cgroup-v2.rst. 1170 1171config CGROUP_DEBUG 1172 bool "Debug controller" 1173 default n 1174 depends on DEBUG_KERNEL 1175 help 1176 This option enables a simple controller that exports 1177 debugging information about the cgroups framework. This 1178 controller is for control cgroup debugging only. Its 1179 interfaces are not stable. 1180 1181 Say N. 1182 1183config SOCK_CGROUP_DATA 1184 bool 1185 default n 1186 1187endif # CGROUPS 1188 1189menuconfig NAMESPACES 1190 bool "Namespaces support" if EXPERT 1191 depends on MULTIUSER 1192 default !EXPERT 1193 help 1194 Provides the way to make tasks work with different objects using 1195 the same id. For example same IPC id may refer to different objects 1196 or same user id or pid may refer to different tasks when used in 1197 different namespaces. 1198 1199if NAMESPACES 1200 1201config UTS_NS 1202 bool "UTS namespace" 1203 default y 1204 help 1205 In this namespace tasks see different info provided with the 1206 uname() system call 1207 1208config TIME_NS 1209 bool "TIME namespace" 1210 depends on GENERIC_VDSO_TIME_NS 1211 default y 1212 help 1213 In this namespace boottime and monotonic clocks can be set. 1214 The time will keep going with the same pace. 1215 1216config IPC_NS 1217 bool "IPC namespace" 1218 depends on (SYSVIPC || POSIX_MQUEUE) 1219 default y 1220 help 1221 In this namespace tasks work with IPC ids which correspond to 1222 different IPC objects in different namespaces. 1223 1224config USER_NS 1225 bool "User namespace" 1226 default n 1227 help 1228 This allows containers, i.e. vservers, to use user namespaces 1229 to provide different user info for different servers. 1230 1231 When user namespaces are enabled in the kernel it is 1232 recommended that the MEMCG option also be enabled and that 1233 user-space use the memory control groups to limit the amount 1234 of memory a memory unprivileged users can use. 1235 1236 If unsure, say N. 1237 1238config PID_NS 1239 bool "PID Namespaces" 1240 default y 1241 help 1242 Support process id namespaces. This allows having multiple 1243 processes with the same pid as long as they are in different 1244 pid namespaces. This is a building block of containers. 1245 1246config NET_NS 1247 bool "Network namespace" 1248 depends on NET 1249 default y 1250 help 1251 Allow user space to create what appear to be multiple instances 1252 of the network stack. 1253 1254endif # NAMESPACES 1255 1256config CHECKPOINT_RESTORE 1257 bool "Checkpoint/restore support" 1258 select PROC_CHILDREN 1259 select KCMP 1260 default n 1261 help 1262 Enables additional kernel features in a sake of checkpoint/restore. 1263 In particular it adds auxiliary prctl codes to setup process text, 1264 data and heap segment sizes, and a few additional /proc filesystem 1265 entries. 1266 1267 If unsure, say N here. 1268 1269config SCHED_AUTOGROUP 1270 bool "Automatic process group scheduling" 1271 select CGROUPS 1272 select CGROUP_SCHED 1273 select FAIR_GROUP_SCHED 1274 help 1275 This option optimizes the scheduler for common desktop workloads by 1276 automatically creating and populating task groups. This separation 1277 of workloads isolates aggressive CPU burners (like build jobs) from 1278 desktop applications. Task group autogeneration is currently based 1279 upon task session. 1280 1281config SYSFS_DEPRECATED 1282 bool "Enable deprecated sysfs features to support old userspace tools" 1283 depends on SYSFS 1284 default n 1285 help 1286 This option adds code that switches the layout of the "block" class 1287 devices, to not show up in /sys/class/block/, but only in 1288 /sys/block/. 1289 1290 This switch is only active when the sysfs.deprecated=1 boot option is 1291 passed or the SYSFS_DEPRECATED_V2 option is set. 1292 1293 This option allows new kernels to run on old distributions and tools, 1294 which might get confused by /sys/class/block/. Since 2007/2008 all 1295 major distributions and tools handle this just fine. 1296 1297 Recent distributions and userspace tools after 2009/2010 depend on 1298 the existence of /sys/class/block/, and will not work with this 1299 option enabled. 1300 1301 Only if you are using a new kernel on an old distribution, you might 1302 need to say Y here. 1303 1304config SYSFS_DEPRECATED_V2 1305 bool "Enable deprecated sysfs features by default" 1306 default n 1307 depends on SYSFS 1308 depends on SYSFS_DEPRECATED 1309 help 1310 Enable deprecated sysfs by default. 1311 1312 See the CONFIG_SYSFS_DEPRECATED option for more details about this 1313 option. 1314 1315 Only if you are using a new kernel on an old distribution, you might 1316 need to say Y here. Even then, odds are you would not need it 1317 enabled, you can always pass the boot option if absolutely necessary. 1318 1319config RELAY 1320 bool "Kernel->user space relay support (formerly relayfs)" 1321 select IRQ_WORK 1322 help 1323 This option enables support for relay interface support in 1324 certain file systems (such as debugfs). 1325 It is designed to provide an efficient mechanism for tools and 1326 facilities to relay large amounts of data from kernel space to 1327 user space. 1328 1329 If unsure, say N. 1330 1331config BLK_DEV_INITRD 1332 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1333 help 1334 The initial RAM filesystem is a ramfs which is loaded by the 1335 boot loader (loadlin or lilo) and that is mounted as root 1336 before the normal boot procedure. It is typically used to 1337 load modules needed to mount the "real" root file system, 1338 etc. See <file:Documentation/admin-guide/initrd.rst> for details. 1339 1340 If RAM disk support (BLK_DEV_RAM) is also included, this 1341 also enables initial RAM disk (initrd) support and adds 1342 15 Kbytes (more on some other architectures) to the kernel size. 1343 1344 If unsure say Y. 1345 1346if BLK_DEV_INITRD 1347 1348source "usr/Kconfig" 1349 1350endif 1351 1352config BOOT_CONFIG 1353 bool "Boot config support" 1354 select BLK_DEV_INITRD 1355 help 1356 Extra boot config allows system admin to pass a config file as 1357 complemental extension of kernel cmdline when booting. 1358 The boot config file must be attached at the end of initramfs 1359 with checksum, size and magic word. 1360 See <file:Documentation/admin-guide/bootconfig.rst> for details. 1361 1362 If unsure, say Y. 1363 1364config INITRAMFS_PRESERVE_MTIME 1365 bool "Preserve cpio archive mtimes in initramfs" 1366 default y 1367 help 1368 Each entry in an initramfs cpio archive carries an mtime value. When 1369 enabled, extracted cpio items take this mtime, with directory mtime 1370 setting deferred until after creation of any child entries. 1371 1372 If unsure, say Y. 1373 1374choice 1375 prompt "Compiler optimization level" 1376 default CC_OPTIMIZE_FOR_PERFORMANCE 1377 1378config CC_OPTIMIZE_FOR_PERFORMANCE 1379 bool "Optimize for performance (-O2)" 1380 help 1381 This is the default optimization level for the kernel, building 1382 with the "-O2" compiler flag for best performance and most 1383 helpful compile-time warnings. 1384 1385config CC_OPTIMIZE_FOR_PERFORMANCE_O3 1386 bool "Optimize more for performance (-O3)" 1387 depends on ARC 1388 help 1389 Choosing this option will pass "-O3" to your compiler to optimize 1390 the kernel yet more for performance. 1391 1392config CC_OPTIMIZE_FOR_SIZE 1393 bool "Optimize for size (-Os)" 1394 help 1395 Choosing this option will pass "-Os" to your compiler resulting 1396 in a smaller kernel. 1397 1398endchoice 1399 1400config HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1401 bool 1402 help 1403 This requires that the arch annotates or otherwise protects 1404 its external entry points from being discarded. Linker scripts 1405 must also merge .text.*, .data.*, and .bss.* correctly into 1406 output sections. Care must be taken not to pull in unrelated 1407 sections (e.g., '.text.init'). Typically '.' in section names 1408 is used to distinguish them from label names / C identifiers. 1409 1410config LD_DEAD_CODE_DATA_ELIMINATION 1411 bool "Dead code and data elimination (EXPERIMENTAL)" 1412 depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1413 depends on EXPERT 1414 depends on $(cc-option,-ffunction-sections -fdata-sections) 1415 depends on $(ld-option,--gc-sections) 1416 help 1417 Enable this if you want to do dead code and data elimination with 1418 the linker by compiling with -ffunction-sections -fdata-sections, 1419 and linking with --gc-sections. 1420 1421 This can reduce on disk and in-memory size of the kernel 1422 code and static data, particularly for small configs and 1423 on small systems. This has the possibility of introducing 1424 silently broken kernel if the required annotations are not 1425 present. This option is not well tested yet, so use at your 1426 own risk. 1427 1428config LD_ORPHAN_WARN 1429 def_bool y 1430 depends on ARCH_WANT_LD_ORPHAN_WARN 1431 depends on $(ld-option,--orphan-handling=warn) 1432 1433config SYSCTL 1434 bool 1435 1436config HAVE_UID16 1437 bool 1438 1439config SYSCTL_EXCEPTION_TRACE 1440 bool 1441 help 1442 Enable support for /proc/sys/debug/exception-trace. 1443 1444config SYSCTL_ARCH_UNALIGN_NO_WARN 1445 bool 1446 help 1447 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap 1448 Allows arch to define/use @no_unaligned_warning to possibly warn 1449 about unaligned access emulation going on under the hood. 1450 1451config SYSCTL_ARCH_UNALIGN_ALLOW 1452 bool 1453 help 1454 Enable support for /proc/sys/kernel/unaligned-trap 1455 Allows arches to define/use @unaligned_enabled to runtime toggle 1456 the unaligned access emulation. 1457 see arch/parisc/kernel/unaligned.c for reference 1458 1459config HAVE_PCSPKR_PLATFORM 1460 bool 1461 1462# interpreter that classic socket filters depend on 1463config BPF 1464 bool 1465 1466menuconfig EXPERT 1467 bool "Configure standard kernel features (expert users)" 1468 # Unhide debug options, to make the on-by-default options visible 1469 select DEBUG_KERNEL 1470 help 1471 This option allows certain base kernel options and settings 1472 to be disabled or tweaked. This is for specialized 1473 environments which can tolerate a "non-standard" kernel. 1474 Only use this if you really know what you are doing. 1475 1476config UID16 1477 bool "Enable 16-bit UID system calls" if EXPERT 1478 depends on HAVE_UID16 && MULTIUSER 1479 default y 1480 help 1481 This enables the legacy 16-bit UID syscall wrappers. 1482 1483config MULTIUSER 1484 bool "Multiple users, groups and capabilities support" if EXPERT 1485 default y 1486 help 1487 This option enables support for non-root users, groups and 1488 capabilities. 1489 1490 If you say N here, all processes will run with UID 0, GID 0, and all 1491 possible capabilities. Saying N here also compiles out support for 1492 system calls related to UIDs, GIDs, and capabilities, such as setuid, 1493 setgid, and capset. 1494 1495 If unsure, say Y here. 1496 1497config SGETMASK_SYSCALL 1498 bool "sgetmask/ssetmask syscalls support" if EXPERT 1499 def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH 1500 help 1501 sys_sgetmask and sys_ssetmask are obsolete system calls 1502 no longer supported in libc but still enabled by default in some 1503 architectures. 1504 1505 If unsure, leave the default option here. 1506 1507config SYSFS_SYSCALL 1508 bool "Sysfs syscall support" if EXPERT 1509 default y 1510 help 1511 sys_sysfs is an obsolete system call no longer supported in libc. 1512 Note that disabling this option is more secure but might break 1513 compatibility with some systems. 1514 1515 If unsure say Y here. 1516 1517config FHANDLE 1518 bool "open by fhandle syscalls" if EXPERT 1519 select EXPORTFS 1520 default y 1521 help 1522 If you say Y here, a user level program will be able to map 1523 file names to handle and then later use the handle for 1524 different file system operations. This is useful in implementing 1525 userspace file servers, which now track files using handles instead 1526 of names. The handle would remain the same even if file names 1527 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 1528 syscalls. 1529 1530config POSIX_TIMERS 1531 bool "Posix Clocks & timers" if EXPERT 1532 default y 1533 help 1534 This includes native support for POSIX timers to the kernel. 1535 Some embedded systems have no use for them and therefore they 1536 can be configured out to reduce the size of the kernel image. 1537 1538 When this option is disabled, the following syscalls won't be 1539 available: timer_create, timer_gettime: timer_getoverrun, 1540 timer_settime, timer_delete, clock_adjtime, getitimer, 1541 setitimer, alarm. Furthermore, the clock_settime, clock_gettime, 1542 clock_getres and clock_nanosleep syscalls will be limited to 1543 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only. 1544 1545 If unsure say y. 1546 1547config PRINTK 1548 default y 1549 bool "Enable support for printk" if EXPERT 1550 select IRQ_WORK 1551 help 1552 This option enables normal printk support. Removing it 1553 eliminates most of the message strings from the kernel image 1554 and makes the kernel more or less silent. As this makes it 1555 very difficult to diagnose system problems, saying N here is 1556 strongly discouraged. 1557 1558config BUG 1559 bool "BUG() support" if EXPERT 1560 default y 1561 help 1562 Disabling this option eliminates support for BUG and WARN, reducing 1563 the size of your kernel image and potentially quietly ignoring 1564 numerous fatal conditions. You should only consider disabling this 1565 option for embedded systems with no facilities for reporting errors. 1566 Just say Y. 1567 1568config ELF_CORE 1569 depends on COREDUMP 1570 default y 1571 bool "Enable ELF core dumps" if EXPERT 1572 help 1573 Enable support for generating core dumps. Disabling saves about 4k. 1574 1575 1576config PCSPKR_PLATFORM 1577 bool "Enable PC-Speaker support" if EXPERT 1578 depends on HAVE_PCSPKR_PLATFORM 1579 select I8253_LOCK 1580 default y 1581 help 1582 This option allows to disable the internal PC-Speaker 1583 support, saving some memory. 1584 1585config BASE_FULL 1586 default y 1587 bool "Enable full-sized data structures for core" if EXPERT 1588 help 1589 Disabling this option reduces the size of miscellaneous core 1590 kernel data structures. This saves memory on small machines, 1591 but may reduce performance. 1592 1593config FUTEX 1594 bool "Enable futex support" if EXPERT 1595 depends on !(SPARC32 && SMP) 1596 default y 1597 imply RT_MUTEXES 1598 help 1599 Disabling this option will cause the kernel to be built without 1600 support for "fast userspace mutexes". The resulting kernel may not 1601 run glibc-based applications correctly. 1602 1603config FUTEX_PI 1604 bool 1605 depends on FUTEX && RT_MUTEXES 1606 default y 1607 1608config EPOLL 1609 bool "Enable eventpoll support" if EXPERT 1610 default y 1611 help 1612 Disabling this option will cause the kernel to be built without 1613 support for epoll family of system calls. 1614 1615config SIGNALFD 1616 bool "Enable signalfd() system call" if EXPERT 1617 default y 1618 help 1619 Enable the signalfd() system call that allows to receive signals 1620 on a file descriptor. 1621 1622 If unsure, say Y. 1623 1624config TIMERFD 1625 bool "Enable timerfd() system call" if EXPERT 1626 default y 1627 help 1628 Enable the timerfd() system call that allows to receive timer 1629 events on a file descriptor. 1630 1631 If unsure, say Y. 1632 1633config EVENTFD 1634 bool "Enable eventfd() system call" if EXPERT 1635 default y 1636 help 1637 Enable the eventfd() system call that allows to receive both 1638 kernel notification (ie. KAIO) or userspace notifications. 1639 1640 If unsure, say Y. 1641 1642config SHMEM 1643 bool "Use full shmem filesystem" if EXPERT 1644 default y 1645 depends on MMU 1646 help 1647 The shmem is an internal filesystem used to manage shared memory. 1648 It is backed by swap and manages resource limits. It is also exported 1649 to userspace as tmpfs if TMPFS is enabled. Disabling this 1650 option replaces shmem and tmpfs with the much simpler ramfs code, 1651 which may be appropriate on small systems without swap. 1652 1653config AIO 1654 bool "Enable AIO support" if EXPERT 1655 default y 1656 help 1657 This option enables POSIX asynchronous I/O which may by used 1658 by some high performance threaded applications. Disabling 1659 this option saves about 7k. 1660 1661config IO_URING 1662 bool "Enable IO uring support" if EXPERT 1663 select IO_WQ 1664 default y 1665 help 1666 This option enables support for the io_uring interface, enabling 1667 applications to submit and complete IO through submission and 1668 completion rings that are shared between the kernel and application. 1669 1670config ADVISE_SYSCALLS 1671 bool "Enable madvise/fadvise syscalls" if EXPERT 1672 default y 1673 help 1674 This option enables the madvise and fadvise syscalls, used by 1675 applications to advise the kernel about their future memory or file 1676 usage, improving performance. If building an embedded system where no 1677 applications use these syscalls, you can disable this option to save 1678 space. 1679 1680config HAVE_ARCH_USERFAULTFD_WP 1681 bool 1682 help 1683 Arch has userfaultfd write protection support 1684 1685config HAVE_ARCH_USERFAULTFD_MINOR 1686 bool 1687 help 1688 Arch has userfaultfd minor fault support 1689 1690config MEMBARRIER 1691 bool "Enable membarrier() system call" if EXPERT 1692 default y 1693 help 1694 Enable the membarrier() system call that allows issuing memory 1695 barriers across all running threads, which can be used to distribute 1696 the cost of user-space memory barriers asymmetrically by transforming 1697 pairs of memory barriers into pairs consisting of membarrier() and a 1698 compiler barrier. 1699 1700 If unsure, say Y. 1701 1702config KALLSYMS 1703 bool "Load all symbols for debugging/ksymoops" if EXPERT 1704 default y 1705 help 1706 Say Y here to let the kernel print out symbolic crash information and 1707 symbolic stack backtraces. This increases the size of the kernel 1708 somewhat, as all symbols have to be loaded into the kernel image. 1709 1710config KALLSYMS_ALL 1711 bool "Include all symbols in kallsyms" 1712 depends on DEBUG_KERNEL && KALLSYMS 1713 help 1714 Normally kallsyms only contains the symbols of functions for nicer 1715 OOPS messages and backtraces (i.e., symbols from the text and inittext 1716 sections). This is sufficient for most cases. And only in very rare 1717 cases (e.g., when a debugger is used) all symbols are required (e.g., 1718 names of variables from the data sections, etc). 1719 1720 This option makes sure that all symbols are loaded into the kernel 1721 image (i.e., symbols from all sections) in cost of increased kernel 1722 size (depending on the kernel configuration, it may be 300KiB or 1723 something like this). 1724 1725 Say N unless you really need all symbols. 1726 1727config KALLSYMS_ABSOLUTE_PERCPU 1728 bool 1729 depends on KALLSYMS 1730 default X86_64 && SMP 1731 1732config KALLSYMS_BASE_RELATIVE 1733 bool 1734 depends on KALLSYMS 1735 default !IA64 1736 help 1737 Instead of emitting them as absolute values in the native word size, 1738 emit the symbol references in the kallsyms table as 32-bit entries, 1739 each containing a relative value in the range [base, base + U32_MAX] 1740 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either 1741 an absolute value in the range [0, S32_MAX] or a relative value in the 1742 range [base, base + S32_MAX], where base is the lowest relative symbol 1743 address encountered in the image. 1744 1745 On 64-bit builds, this reduces the size of the address table by 50%, 1746 but more importantly, it results in entries whose values are build 1747 time constants, and no relocation pass is required at runtime to fix 1748 up the entries based on the runtime load address of the kernel. 1749 1750# end of the "standard kernel features (expert users)" menu 1751 1752# syscall, maps, verifier 1753 1754config USERFAULTFD 1755 bool "Enable userfaultfd() system call" 1756 depends on MMU 1757 help 1758 Enable the userfaultfd() system call that allows to intercept and 1759 handle page faults in userland. 1760 1761config ARCH_HAS_MEMBARRIER_CALLBACKS 1762 bool 1763 1764config ARCH_HAS_MEMBARRIER_SYNC_CORE 1765 bool 1766 1767config KCMP 1768 bool "Enable kcmp() system call" if EXPERT 1769 help 1770 Enable the kernel resource comparison system call. It provides 1771 user-space with the ability to compare two processes to see if they 1772 share a common resource, such as a file descriptor or even virtual 1773 memory space. 1774 1775 If unsure, say N. 1776 1777config RSEQ 1778 bool "Enable rseq() system call" if EXPERT 1779 default y 1780 depends on HAVE_RSEQ 1781 select MEMBARRIER 1782 help 1783 Enable the restartable sequences system call. It provides a 1784 user-space cache for the current CPU number value, which 1785 speeds up getting the current CPU number from user-space, 1786 as well as an ABI to speed up user-space operations on 1787 per-CPU data. 1788 1789 If unsure, say Y. 1790 1791config DEBUG_RSEQ 1792 default n 1793 bool "Enabled debugging of rseq() system call" if EXPERT 1794 depends on RSEQ && DEBUG_KERNEL 1795 help 1796 Enable extra debugging checks for the rseq system call. 1797 1798 If unsure, say N. 1799 1800config EMBEDDED 1801 bool "Embedded system" 1802 select EXPERT 1803 help 1804 This option should be enabled if compiling the kernel for 1805 an embedded system so certain expert options are available 1806 for configuration. 1807 1808config HAVE_PERF_EVENTS 1809 bool 1810 help 1811 See tools/perf/design.txt for details. 1812 1813config GUEST_PERF_EVENTS 1814 bool 1815 depends on HAVE_PERF_EVENTS 1816 1817config PERF_USE_VMALLOC 1818 bool 1819 help 1820 See tools/perf/design.txt for details 1821 1822config PC104 1823 bool "PC/104 support" if EXPERT 1824 help 1825 Expose PC/104 form factor device drivers and options available for 1826 selection and configuration. Enable this option if your target 1827 machine has a PC/104 bus. 1828 1829menu "Kernel Performance Events And Counters" 1830 1831config PERF_EVENTS 1832 bool "Kernel performance events and counters" 1833 default y if PROFILING 1834 depends on HAVE_PERF_EVENTS 1835 select IRQ_WORK 1836 select SRCU 1837 help 1838 Enable kernel support for various performance events provided 1839 by software and hardware. 1840 1841 Software events are supported either built-in or via the 1842 use of generic tracepoints. 1843 1844 Most modern CPUs support performance events via performance 1845 counter registers. These registers count the number of certain 1846 types of hw events: such as instructions executed, cachemisses 1847 suffered, or branches mis-predicted - without slowing down the 1848 kernel or applications. These registers can also trigger interrupts 1849 when a threshold number of events have passed - and can thus be 1850 used to profile the code that runs on that CPU. 1851 1852 The Linux Performance Event subsystem provides an abstraction of 1853 these software and hardware event capabilities, available via a 1854 system call and used by the "perf" utility in tools/perf/. It 1855 provides per task and per CPU counters, and it provides event 1856 capabilities on top of those. 1857 1858 Say Y if unsure. 1859 1860config DEBUG_PERF_USE_VMALLOC 1861 default n 1862 bool "Debug: use vmalloc to back perf mmap() buffers" 1863 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC 1864 select PERF_USE_VMALLOC 1865 help 1866 Use vmalloc memory to back perf mmap() buffers. 1867 1868 Mostly useful for debugging the vmalloc code on platforms 1869 that don't require it. 1870 1871 Say N if unsure. 1872 1873endmenu 1874 1875config VM_EVENT_COUNTERS 1876 default y 1877 bool "Enable VM event counters for /proc/vmstat" if EXPERT 1878 help 1879 VM event counters are needed for event counts to be shown. 1880 This option allows the disabling of the VM event counters 1881 on EXPERT systems. /proc/vmstat will only show page counts 1882 if VM event counters are disabled. 1883 1884config SLUB_DEBUG 1885 default y 1886 bool "Enable SLUB debugging support" if EXPERT 1887 depends on SLUB && SYSFS 1888 help 1889 SLUB has extensive debug support features. Disabling these can 1890 result in significant savings in code size. This also disables 1891 SLUB sysfs support. /sys/slab will not exist and there will be 1892 no support for cache validation etc. 1893 1894config COMPAT_BRK 1895 bool "Disable heap randomization" 1896 default y 1897 help 1898 Randomizing heap placement makes heap exploits harder, but it 1899 also breaks ancient binaries (including anything libc5 based). 1900 This option changes the bootup default to heap randomization 1901 disabled, and can be overridden at runtime by setting 1902 /proc/sys/kernel/randomize_va_space to 2. 1903 1904 On non-ancient distros (post-2000 ones) N is usually a safe choice. 1905 1906choice 1907 prompt "Choose SLAB allocator" 1908 default SLUB 1909 help 1910 This option allows to select a slab allocator. 1911 1912config SLAB 1913 bool "SLAB" 1914 depends on !PREEMPT_RT 1915 select HAVE_HARDENED_USERCOPY_ALLOCATOR 1916 help 1917 The regular slab allocator that is established and known to work 1918 well in all environments. It organizes cache hot objects in 1919 per cpu and per node queues. 1920 1921config SLUB 1922 bool "SLUB (Unqueued Allocator)" 1923 select HAVE_HARDENED_USERCOPY_ALLOCATOR 1924 help 1925 SLUB is a slab allocator that minimizes cache line usage 1926 instead of managing queues of cached objects (SLAB approach). 1927 Per cpu caching is realized using slabs of objects instead 1928 of queues of objects. SLUB can use memory efficiently 1929 and has enhanced diagnostics. SLUB is the default choice for 1930 a slab allocator. 1931 1932config SLOB 1933 depends on EXPERT 1934 bool "SLOB (Simple Allocator)" 1935 depends on !PREEMPT_RT 1936 help 1937 SLOB replaces the stock allocator with a drastically simpler 1938 allocator. SLOB is generally more space efficient but 1939 does not perform as well on large systems. 1940 1941endchoice 1942 1943config SLAB_MERGE_DEFAULT 1944 bool "Allow slab caches to be merged" 1945 default y 1946 depends on SLAB || SLUB 1947 help 1948 For reduced kernel memory fragmentation, slab caches can be 1949 merged when they share the same size and other characteristics. 1950 This carries a risk of kernel heap overflows being able to 1951 overwrite objects from merged caches (and more easily control 1952 cache layout), which makes such heap attacks easier to exploit 1953 by attackers. By keeping caches unmerged, these kinds of exploits 1954 can usually only damage objects in the same cache. To disable 1955 merging at runtime, "slab_nomerge" can be passed on the kernel 1956 command line. 1957 1958config SLAB_FREELIST_RANDOM 1959 bool "Randomize slab freelist" 1960 depends on SLAB || SLUB 1961 help 1962 Randomizes the freelist order used on creating new pages. This 1963 security feature reduces the predictability of the kernel slab 1964 allocator against heap overflows. 1965 1966config SLAB_FREELIST_HARDENED 1967 bool "Harden slab freelist metadata" 1968 depends on SLAB || SLUB 1969 help 1970 Many kernel heap attacks try to target slab cache metadata and 1971 other infrastructure. This options makes minor performance 1972 sacrifices to harden the kernel slab allocator against common 1973 freelist exploit methods. Some slab implementations have more 1974 sanity-checking than others. This option is most effective with 1975 CONFIG_SLUB. 1976 1977config SHUFFLE_PAGE_ALLOCATOR 1978 bool "Page allocator randomization" 1979 default SLAB_FREELIST_RANDOM && ACPI_NUMA 1980 help 1981 Randomization of the page allocator improves the average 1982 utilization of a direct-mapped memory-side-cache. See section 1983 5.2.27 Heterogeneous Memory Attribute Table (HMAT) in the ACPI 1984 6.2a specification for an example of how a platform advertises 1985 the presence of a memory-side-cache. There are also incidental 1986 security benefits as it reduces the predictability of page 1987 allocations to compliment SLAB_FREELIST_RANDOM, but the 1988 default granularity of shuffling on the "MAX_ORDER - 1" i.e, 1989 10th order of pages is selected based on cache utilization 1990 benefits on x86. 1991 1992 While the randomization improves cache utilization it may 1993 negatively impact workloads on platforms without a cache. For 1994 this reason, by default, the randomization is enabled only 1995 after runtime detection of a direct-mapped memory-side-cache. 1996 Otherwise, the randomization may be force enabled with the 1997 'page_alloc.shuffle' kernel command line parameter. 1998 1999 Say Y if unsure. 2000 2001config SLUB_CPU_PARTIAL 2002 default y 2003 depends on SLUB && SMP 2004 bool "SLUB per cpu partial cache" 2005 help 2006 Per cpu partial caches accelerate objects allocation and freeing 2007 that is local to a processor at the price of more indeterminism 2008 in the latency of the free. On overflow these caches will be cleared 2009 which requires the taking of locks that may cause latency spikes. 2010 Typically one would choose no for a realtime system. 2011 2012config MMAP_ALLOW_UNINITIALIZED 2013 bool "Allow mmapped anonymous memory to be uninitialized" 2014 depends on EXPERT && !MMU 2015 default n 2016 help 2017 Normally, and according to the Linux spec, anonymous memory obtained 2018 from mmap() has its contents cleared before it is passed to 2019 userspace. Enabling this config option allows you to request that 2020 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus 2021 providing a huge performance boost. If this option is not enabled, 2022 then the flag will be ignored. 2023 2024 This is taken advantage of by uClibc's malloc(), and also by 2025 ELF-FDPIC binfmt's brk and stack allocator. 2026 2027 Because of the obvious security issues, this option should only be 2028 enabled on embedded devices where you control what is run in 2029 userspace. Since that isn't generally a problem on no-MMU systems, 2030 it is normally safe to say Y here. 2031 2032 See Documentation/admin-guide/mm/nommu-mmap.rst for more information. 2033 2034config SYSTEM_DATA_VERIFICATION 2035 def_bool n 2036 select SYSTEM_TRUSTED_KEYRING 2037 select KEYS 2038 select CRYPTO 2039 select CRYPTO_RSA 2040 select ASYMMETRIC_KEY_TYPE 2041 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE 2042 select ASN1 2043 select OID_REGISTRY 2044 select X509_CERTIFICATE_PARSER 2045 select PKCS7_MESSAGE_PARSER 2046 help 2047 Provide PKCS#7 message verification using the contents of the system 2048 trusted keyring to provide public keys. This then can be used for 2049 module verification, kexec image verification and firmware blob 2050 verification. 2051 2052config PROFILING 2053 bool "Profiling support" 2054 help 2055 Say Y here to enable the extended profiling support mechanisms used 2056 by profilers. 2057 2058# 2059# Place an empty function call at each tracepoint site. Can be 2060# dynamically changed for a probe function. 2061# 2062config TRACEPOINTS 2063 bool 2064 2065endmenu # General setup 2066 2067source "arch/Kconfig" 2068 2069config RT_MUTEXES 2070 bool 2071 default y if PREEMPT_RT 2072 2073config BASE_SMALL 2074 int 2075 default 0 if BASE_FULL 2076 default 1 if !BASE_FULL 2077 2078config MODULE_SIG_FORMAT 2079 def_bool n 2080 select SYSTEM_DATA_VERIFICATION 2081 2082menuconfig MODULES 2083 bool "Enable loadable module support" 2084 modules 2085 help 2086 Kernel modules are small pieces of compiled code which can 2087 be inserted in the running kernel, rather than being 2088 permanently built into the kernel. You use the "modprobe" 2089 tool to add (and sometimes remove) them. If you say Y here, 2090 many parts of the kernel can be built as modules (by 2091 answering M instead of Y where indicated): this is most 2092 useful for infrequently used options which are not required 2093 for booting. For more information, see the man pages for 2094 modprobe, lsmod, modinfo, insmod and rmmod. 2095 2096 If you say Y here, you will need to run "make 2097 modules_install" to put the modules under /lib/modules/ 2098 where modprobe can find them (you may need to be root to do 2099 this). 2100 2101 If unsure, say Y. 2102 2103if MODULES 2104 2105config MODULE_FORCE_LOAD 2106 bool "Forced module loading" 2107 default n 2108 help 2109 Allow loading of modules without version information (ie. modprobe 2110 --force). Forced module loading sets the 'F' (forced) taint flag and 2111 is usually a really bad idea. 2112 2113config MODULE_UNLOAD 2114 bool "Module unloading" 2115 help 2116 Without this option you will not be able to unload any 2117 modules (note that some modules may not be unloadable 2118 anyway), which makes your kernel smaller, faster 2119 and simpler. If unsure, say Y. 2120 2121config MODULE_FORCE_UNLOAD 2122 bool "Forced module unloading" 2123 depends on MODULE_UNLOAD 2124 help 2125 This option allows you to force a module to unload, even if the 2126 kernel believes it is unsafe: the kernel will remove the module 2127 without waiting for anyone to stop using it (using the -f option to 2128 rmmod). This is mainly for kernel developers and desperate users. 2129 If unsure, say N. 2130 2131config MODVERSIONS 2132 bool "Module versioning support" 2133 help 2134 Usually, you have to use modules compiled with your kernel. 2135 Saying Y here makes it sometimes possible to use modules 2136 compiled for different kernels, by adding enough information 2137 to the modules to (hopefully) spot any changes which would 2138 make them incompatible with the kernel you are running. If 2139 unsure, say N. 2140 2141config ASM_MODVERSIONS 2142 bool 2143 default HAVE_ASM_MODVERSIONS && MODVERSIONS 2144 help 2145 This enables module versioning for exported symbols also from 2146 assembly. This can be enabled only when the target architecture 2147 supports it. 2148 2149config MODULE_REL_CRCS 2150 bool 2151 depends on MODVERSIONS 2152 2153config MODULE_SRCVERSION_ALL 2154 bool "Source checksum for all modules" 2155 help 2156 Modules which contain a MODULE_VERSION get an extra "srcversion" 2157 field inserted into their modinfo section, which contains a 2158 sum of the source files which made it. This helps maintainers 2159 see exactly which source was used to build a module (since 2160 others sometimes change the module source without updating 2161 the version). With this option, such a "srcversion" field 2162 will be created for all modules. If unsure, say N. 2163 2164config MODULE_SIG 2165 bool "Module signature verification" 2166 select MODULE_SIG_FORMAT 2167 help 2168 Check modules for valid signatures upon load: the signature 2169 is simply appended to the module. For more information see 2170 <file:Documentation/admin-guide/module-signing.rst>. 2171 2172 Note that this option adds the OpenSSL development packages as a 2173 kernel build dependency so that the signing tool can use its crypto 2174 library. 2175 2176 You should enable this option if you wish to use either 2177 CONFIG_SECURITY_LOCKDOWN_LSM or lockdown functionality imposed via 2178 another LSM - otherwise unsigned modules will be loadable regardless 2179 of the lockdown policy. 2180 2181 !!!WARNING!!! If you enable this option, you MUST make sure that the 2182 module DOES NOT get stripped after being signed. This includes the 2183 debuginfo strip done by some packagers (such as rpmbuild) and 2184 inclusion into an initramfs that wants the module size reduced. 2185 2186config MODULE_SIG_FORCE 2187 bool "Require modules to be validly signed" 2188 depends on MODULE_SIG 2189 help 2190 Reject unsigned modules or signed modules for which we don't have a 2191 key. Without this, such modules will simply taint the kernel. 2192 2193config MODULE_SIG_ALL 2194 bool "Automatically sign all modules" 2195 default y 2196 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2197 help 2198 Sign all modules during make modules_install. Without this option, 2199 modules must be signed manually, using the scripts/sign-file tool. 2200 2201comment "Do not forget to sign required modules with scripts/sign-file" 2202 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL 2203 2204choice 2205 prompt "Which hash algorithm should modules be signed with?" 2206 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2207 help 2208 This determines which sort of hashing algorithm will be used during 2209 signature generation. This algorithm _must_ be built into the kernel 2210 directly so that signature verification can take place. It is not 2211 possible to load a signed module containing the algorithm to check 2212 the signature on that module. 2213 2214config MODULE_SIG_SHA1 2215 bool "Sign modules with SHA-1" 2216 select CRYPTO_SHA1 2217 2218config MODULE_SIG_SHA224 2219 bool "Sign modules with SHA-224" 2220 select CRYPTO_SHA256 2221 2222config MODULE_SIG_SHA256 2223 bool "Sign modules with SHA-256" 2224 select CRYPTO_SHA256 2225 2226config MODULE_SIG_SHA384 2227 bool "Sign modules with SHA-384" 2228 select CRYPTO_SHA512 2229 2230config MODULE_SIG_SHA512 2231 bool "Sign modules with SHA-512" 2232 select CRYPTO_SHA512 2233 2234endchoice 2235 2236config MODULE_SIG_HASH 2237 string 2238 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2239 default "sha1" if MODULE_SIG_SHA1 2240 default "sha224" if MODULE_SIG_SHA224 2241 default "sha256" if MODULE_SIG_SHA256 2242 default "sha384" if MODULE_SIG_SHA384 2243 default "sha512" if MODULE_SIG_SHA512 2244 2245choice 2246 prompt "Module compression mode" 2247 help 2248 This option allows you to choose the algorithm which will be used to 2249 compress modules when 'make modules_install' is run. (or, you can 2250 choose to not compress modules at all.) 2251 2252 External modules will also be compressed in the same way during the 2253 installation. 2254 2255 For modules inside an initrd or initramfs, it's more efficient to 2256 compress the whole initrd or initramfs instead. 2257 2258 This is fully compatible with signed modules. 2259 2260 Please note that the tool used to load modules needs to support the 2261 corresponding algorithm. module-init-tools MAY support gzip, and kmod 2262 MAY support gzip, xz and zstd. 2263 2264 Your build system needs to provide the appropriate compression tool 2265 to compress the modules. 2266 2267 If in doubt, select 'None'. 2268 2269config MODULE_COMPRESS_NONE 2270 bool "None" 2271 help 2272 Do not compress modules. The installed modules are suffixed 2273 with .ko. 2274 2275config MODULE_COMPRESS_GZIP 2276 bool "GZIP" 2277 help 2278 Compress modules with GZIP. The installed modules are suffixed 2279 with .ko.gz. 2280 2281config MODULE_COMPRESS_XZ 2282 bool "XZ" 2283 help 2284 Compress modules with XZ. The installed modules are suffixed 2285 with .ko.xz. 2286 2287config MODULE_COMPRESS_ZSTD 2288 bool "ZSTD" 2289 help 2290 Compress modules with ZSTD. The installed modules are suffixed 2291 with .ko.zst. 2292 2293endchoice 2294 2295config MODULE_DECOMPRESS 2296 bool "Support in-kernel module decompression" 2297 depends on MODULE_COMPRESS_GZIP || MODULE_COMPRESS_XZ 2298 select ZLIB_INFLATE if MODULE_COMPRESS_GZIP 2299 select XZ_DEC if MODULE_COMPRESS_XZ 2300 help 2301 2302 Support for decompressing kernel modules by the kernel itself 2303 instead of relying on userspace to perform this task. Useful when 2304 load pinning security policy is enabled. 2305 2306 If unsure, say N. 2307 2308config MODULE_ALLOW_MISSING_NAMESPACE_IMPORTS 2309 bool "Allow loading of modules with missing namespace imports" 2310 help 2311 Symbols exported with EXPORT_SYMBOL_NS*() are considered exported in 2312 a namespace. A module that makes use of a symbol exported with such a 2313 namespace is required to import the namespace via MODULE_IMPORT_NS(). 2314 There is no technical reason to enforce correct namespace imports, 2315 but it creates consistency between symbols defining namespaces and 2316 users importing namespaces they make use of. This option relaxes this 2317 requirement and lifts the enforcement when loading a module. 2318 2319 If unsure, say N. 2320 2321config MODPROBE_PATH 2322 string "Path to modprobe binary" 2323 default "/sbin/modprobe" 2324 help 2325 When kernel code requests a module, it does so by calling 2326 the "modprobe" userspace utility. This option allows you to 2327 set the path where that binary is found. This can be changed 2328 at runtime via the sysctl file 2329 /proc/sys/kernel/modprobe. Setting this to the empty string 2330 removes the kernel's ability to request modules (but 2331 userspace can still load modules explicitly). 2332 2333config TRIM_UNUSED_KSYMS 2334 bool "Trim unused exported kernel symbols" if EXPERT 2335 depends on !COMPILE_TEST 2336 help 2337 The kernel and some modules make many symbols available for 2338 other modules to use via EXPORT_SYMBOL() and variants. Depending 2339 on the set of modules being selected in your kernel configuration, 2340 many of those exported symbols might never be used. 2341 2342 This option allows for unused exported symbols to be dropped from 2343 the build. In turn, this provides the compiler more opportunities 2344 (especially when using LTO) for optimizing the code and reducing 2345 binary size. This might have some security advantages as well. 2346 2347 If unsure, or if you need to build out-of-tree modules, say N. 2348 2349config UNUSED_KSYMS_WHITELIST 2350 string "Whitelist of symbols to keep in ksymtab" 2351 depends on TRIM_UNUSED_KSYMS 2352 help 2353 By default, all unused exported symbols will be un-exported from the 2354 build when TRIM_UNUSED_KSYMS is selected. 2355 2356 UNUSED_KSYMS_WHITELIST allows to whitelist symbols that must be kept 2357 exported at all times, even in absence of in-tree users. The value to 2358 set here is the path to a text file containing the list of symbols, 2359 one per line. The path can be absolute, or relative to the kernel 2360 source tree. 2361 2362endif # MODULES 2363 2364config MODULES_TREE_LOOKUP 2365 def_bool y 2366 depends on PERF_EVENTS || TRACING || CFI_CLANG 2367 2368config INIT_ALL_POSSIBLE 2369 bool 2370 help 2371 Back when each arch used to define their own cpu_online_mask and 2372 cpu_possible_mask, some of them chose to initialize cpu_possible_mask 2373 with all 1s, and others with all 0s. When they were centralised, 2374 it was better to provide this option than to break all the archs 2375 and have several arch maintainers pursuing me down dark alleys. 2376 2377source "block/Kconfig" 2378 2379config PREEMPT_NOTIFIERS 2380 bool 2381 2382config PADATA 2383 depends on SMP 2384 bool 2385 2386config ASN1 2387 tristate 2388 help 2389 Build a simple ASN.1 grammar compiler that produces a bytecode output 2390 that can be interpreted by the ASN.1 stream decoder and used to 2391 inform it as to what tags are to be expected in a stream and what 2392 functions to call on what tags. 2393 2394source "kernel/Kconfig.locks" 2395 2396config ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE 2397 bool 2398 2399config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE 2400 bool 2401 2402# It may be useful for an architecture to override the definitions of the 2403# SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h> 2404# and the COMPAT_ variants in <linux/compat.h>, in particular to use a 2405# different calling convention for syscalls. They can also override the 2406# macros for not-implemented syscalls in kernel/sys_ni.c and 2407# kernel/time/posix-stubs.c. All these overrides need to be available in 2408# <asm/syscall_wrapper.h>. 2409config ARCH_HAS_SYSCALL_WRAPPER 2410 def_bool n 2411