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) $(m64-flag)) if 64BIT 66 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(m32-flag)) 67 68config CC_CAN_LINK_STATIC 69 bool 70 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(m64-flag) -static) if 64BIT 71 default $(success,$(srctree)/scripts/cc-can-link.sh $(CC) $(CLANG_FLAGS) $(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 CONSTRUCTORS 90 bool 91 92config IRQ_WORK 93 bool 94 95config BUILDTIME_TABLE_SORT 96 bool 97 98config THREAD_INFO_IN_TASK 99 bool 100 help 101 Select this to move thread_info off the stack into task_struct. To 102 make this work, an arch will need to remove all thread_info fields 103 except flags and fix any runtime bugs. 104 105 One subtle change that will be needed is to use try_get_task_stack() 106 and put_task_stack() in save_thread_stack_tsk() and get_wchan(). 107 108menu "General setup" 109 110config BROKEN 111 bool 112 113config BROKEN_ON_SMP 114 bool 115 depends on BROKEN || !SMP 116 default y 117 118config INIT_ENV_ARG_LIMIT 119 int 120 default 32 if !UML 121 default 128 if UML 122 help 123 Maximum of each of the number of arguments and environment 124 variables passed to init from the kernel command line. 125 126config COMPILE_TEST 127 bool "Compile also drivers which will not load" 128 depends on HAS_IOMEM 129 help 130 Some drivers can be compiled on a different platform than they are 131 intended to be run on. Despite they cannot be loaded there (or even 132 when they load they cannot be used due to missing HW support), 133 developers still, opposing to distributors, might want to build such 134 drivers to compile-test them. 135 136 If you are a developer and want to build everything available, say Y 137 here. If you are a user/distributor, say N here to exclude useless 138 drivers to be distributed. 139 140config UAPI_HEADER_TEST 141 bool "Compile test UAPI headers" 142 depends on HEADERS_INSTALL && CC_CAN_LINK 143 help 144 Compile test headers exported to user-space to ensure they are 145 self-contained, i.e. compilable as standalone units. 146 147 If you are a developer or tester and want to ensure the exported 148 headers are self-contained, say Y here. Otherwise, choose N. 149 150config LOCALVERSION 151 string "Local version - append to kernel release" 152 help 153 Append an extra string to the end of your kernel version. 154 This will show up when you type uname, for example. 155 The string you set here will be appended after the contents of 156 any files with a filename matching localversion* in your 157 object and source tree, in that order. Your total string can 158 be a maximum of 64 characters. 159 160config LOCALVERSION_AUTO 161 bool "Automatically append version information to the version string" 162 default y 163 depends on !COMPILE_TEST 164 help 165 This will try to automatically determine if the current tree is a 166 release tree by looking for git tags that belong to the current 167 top of tree revision. 168 169 A string of the format -gxxxxxxxx will be added to the localversion 170 if a git-based tree is found. The string generated by this will be 171 appended after any matching localversion* files, and after the value 172 set in CONFIG_LOCALVERSION. 173 174 (The actual string used here is the first eight characters produced 175 by running the command: 176 177 $ git rev-parse --verify HEAD 178 179 which is done within the script "scripts/setlocalversion".) 180 181config BUILD_SALT 182 string "Build ID Salt" 183 default "" 184 help 185 The build ID is used to link binaries and their debug info. Setting 186 this option will use the value in the calculation of the build id. 187 This is mostly useful for distributions which want to ensure the 188 build is unique between builds. It's safe to leave the default. 189 190config HAVE_KERNEL_GZIP 191 bool 192 193config HAVE_KERNEL_BZIP2 194 bool 195 196config HAVE_KERNEL_LZMA 197 bool 198 199config HAVE_KERNEL_XZ 200 bool 201 202config HAVE_KERNEL_LZO 203 bool 204 205config HAVE_KERNEL_LZ4 206 bool 207 208config HAVE_KERNEL_ZSTD 209 bool 210 211config HAVE_KERNEL_UNCOMPRESSED 212 bool 213 214choice 215 prompt "Kernel compression mode" 216 default KERNEL_GZIP 217 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 218 help 219 The linux kernel is a kind of self-extracting executable. 220 Several compression algorithms are available, which differ 221 in efficiency, compression and decompression speed. 222 Compression speed is only relevant when building a kernel. 223 Decompression speed is relevant at each boot. 224 225 If you have any problems with bzip2 or lzma compressed 226 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 227 version of this functionality (bzip2 only), for 2.4, was 228 supplied by Christian Ludwig) 229 230 High compression options are mostly useful for users, who 231 are low on disk space (embedded systems), but for whom ram 232 size matters less. 233 234 If in doubt, select 'gzip' 235 236config KERNEL_GZIP 237 bool "Gzip" 238 depends on HAVE_KERNEL_GZIP 239 help 240 The old and tried gzip compression. It provides a good balance 241 between compression ratio and decompression speed. 242 243config KERNEL_BZIP2 244 bool "Bzip2" 245 depends on HAVE_KERNEL_BZIP2 246 help 247 Its compression ratio and speed is intermediate. 248 Decompression speed is slowest among the choices. The kernel 249 size is about 10% smaller with bzip2, in comparison to gzip. 250 Bzip2 uses a large amount of memory. For modern kernels you 251 will need at least 8MB RAM or more for booting. 252 253config KERNEL_LZMA 254 bool "LZMA" 255 depends on HAVE_KERNEL_LZMA 256 help 257 This compression algorithm's ratio is best. Decompression speed 258 is between gzip and bzip2. Compression is slowest. 259 The kernel size is about 33% smaller with LZMA in comparison to gzip. 260 261config KERNEL_XZ 262 bool "XZ" 263 depends on HAVE_KERNEL_XZ 264 help 265 XZ uses the LZMA2 algorithm and instruction set specific 266 BCJ filters which can improve compression ratio of executable 267 code. The size of the kernel is about 30% smaller with XZ in 268 comparison to gzip. On architectures for which there is a BCJ 269 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 270 will create a few percent smaller kernel than plain LZMA. 271 272 The speed is about the same as with LZMA: The decompression 273 speed of XZ is better than that of bzip2 but worse than gzip 274 and LZO. Compression is slow. 275 276config KERNEL_LZO 277 bool "LZO" 278 depends on HAVE_KERNEL_LZO 279 help 280 Its compression ratio is the poorest among the choices. The kernel 281 size is about 10% bigger than gzip; however its speed 282 (both compression and decompression) is the fastest. 283 284config KERNEL_LZ4 285 bool "LZ4" 286 depends on HAVE_KERNEL_LZ4 287 help 288 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding. 289 A preliminary version of LZ4 de/compression tool is available at 290 <https://code.google.com/p/lz4/>. 291 292 Its compression ratio is worse than LZO. The size of the kernel 293 is about 8% bigger than LZO. But the decompression speed is 294 faster than LZO. 295 296config KERNEL_ZSTD 297 bool "ZSTD" 298 depends on HAVE_KERNEL_ZSTD 299 help 300 ZSTD is a compression algorithm targeting intermediate compression 301 with fast decompression speed. It will compress better than GZIP and 302 decompress around the same speed as LZO, but slower than LZ4. You 303 will need at least 192 KB RAM or more for booting. The zstd command 304 line tool is required for compression. 305 306config KERNEL_UNCOMPRESSED 307 bool "None" 308 depends on HAVE_KERNEL_UNCOMPRESSED 309 help 310 Produce uncompressed kernel image. This option is usually not what 311 you want. It is useful for debugging the kernel in slow simulation 312 environments, where decompressing and moving the kernel is awfully 313 slow. This option allows early boot code to skip the decompressor 314 and jump right at uncompressed kernel image. 315 316endchoice 317 318config DEFAULT_INIT 319 string "Default init path" 320 default "" 321 help 322 This option determines the default init for the system if no init= 323 option is passed on the kernel command line. If the requested path is 324 not present, we will still then move on to attempting further 325 locations (e.g. /sbin/init, etc). If this is empty, we will just use 326 the fallback list when init= is not passed. 327 328config DEFAULT_HOSTNAME 329 string "Default hostname" 330 default "(none)" 331 help 332 This option determines the default system hostname before userspace 333 calls sethostname(2). The kernel traditionally uses "(none)" here, 334 but you may wish to use a different default here to make a minimal 335 system more usable with less configuration. 336 337# 338# For some reason microblaze and nios2 hard code SWAP=n. Hopefully we can 339# add proper SWAP support to them, in which case this can be remove. 340# 341config ARCH_NO_SWAP 342 bool 343 344config SWAP 345 bool "Support for paging of anonymous memory (swap)" 346 depends on MMU && BLOCK && !ARCH_NO_SWAP 347 default y 348 help 349 This option allows you to choose whether you want to have support 350 for so called swap devices or swap files in your kernel that are 351 used to provide more virtual memory than the actual RAM present 352 in your computer. If unsure say Y. 353 354config SYSVIPC 355 bool "System V IPC" 356 help 357 Inter Process Communication is a suite of library functions and 358 system calls which let processes (running programs) synchronize and 359 exchange information. It is generally considered to be a good thing, 360 and some programs won't run unless you say Y here. In particular, if 361 you want to run the DOS emulator dosemu under Linux (read the 362 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 363 you'll need to say Y here. 364 365 You can find documentation about IPC with "info ipc" and also in 366 section 6.4 of the Linux Programmer's Guide, available from 367 <http://www.tldp.org/guides.html>. 368 369config SYSVIPC_SYSCTL 370 bool 371 depends on SYSVIPC 372 depends on SYSCTL 373 default y 374 375config POSIX_MQUEUE 376 bool "POSIX Message Queues" 377 depends on NET 378 help 379 POSIX variant of message queues is a part of IPC. In POSIX message 380 queues every message has a priority which decides about succession 381 of receiving it by a process. If you want to compile and run 382 programs written e.g. for Solaris with use of its POSIX message 383 queues (functions mq_*) say Y here. 384 385 POSIX message queues are visible as a filesystem called 'mqueue' 386 and can be mounted somewhere if you want to do filesystem 387 operations on message queues. 388 389 If unsure, say Y. 390 391config POSIX_MQUEUE_SYSCTL 392 bool 393 depends on POSIX_MQUEUE 394 depends on SYSCTL 395 default y 396 397config WATCH_QUEUE 398 bool "General notification queue" 399 default n 400 help 401 402 This is a general notification queue for the kernel to pass events to 403 userspace by splicing them into pipes. It can be used in conjunction 404 with watches for key/keyring change notifications and device 405 notifications. 406 407 See Documentation/watch_queue.rst 408 409config CROSS_MEMORY_ATTACH 410 bool "Enable process_vm_readv/writev syscalls" 411 depends on MMU 412 default y 413 help 414 Enabling this option adds the system calls process_vm_readv and 415 process_vm_writev which allow a process with the correct privileges 416 to directly read from or write to another process' address space. 417 See the man page for more details. 418 419config USELIB 420 bool "uselib syscall" 421 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION 422 help 423 This option enables the uselib syscall, a system call used in the 424 dynamic linker from libc5 and earlier. glibc does not use this 425 system call. If you intend to run programs built on libc5 or 426 earlier, you may need to enable this syscall. Current systems 427 running glibc can safely disable this. 428 429config AUDIT 430 bool "Auditing support" 431 depends on NET 432 help 433 Enable auditing infrastructure that can be used with another 434 kernel subsystem, such as SELinux (which requires this for 435 logging of avc messages output). System call auditing is included 436 on architectures which support it. 437 438config HAVE_ARCH_AUDITSYSCALL 439 bool 440 441config AUDITSYSCALL 442 def_bool y 443 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL 444 select FSNOTIFY 445 446source "kernel/irq/Kconfig" 447source "kernel/time/Kconfig" 448source "kernel/bpf/Kconfig" 449source "kernel/Kconfig.preempt" 450 451menu "CPU/Task time and stats accounting" 452 453config VIRT_CPU_ACCOUNTING 454 bool 455 456choice 457 prompt "Cputime accounting" 458 default TICK_CPU_ACCOUNTING if !PPC64 459 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64 460 461# Kind of a stub config for the pure tick based cputime accounting 462config TICK_CPU_ACCOUNTING 463 bool "Simple tick based cputime accounting" 464 depends on !S390 && !NO_HZ_FULL 465 help 466 This is the basic tick based cputime accounting that maintains 467 statistics about user, system and idle time spent on per jiffies 468 granularity. 469 470 If unsure, say Y. 471 472config VIRT_CPU_ACCOUNTING_NATIVE 473 bool "Deterministic task and CPU time accounting" 474 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL 475 select VIRT_CPU_ACCOUNTING 476 help 477 Select this option to enable more accurate task and CPU time 478 accounting. This is done by reading a CPU counter on each 479 kernel entry and exit and on transitions within the kernel 480 between system, softirq and hardirq state, so there is a 481 small performance impact. In the case of s390 or IBM POWER > 5, 482 this also enables accounting of stolen time on logically-partitioned 483 systems. 484 485config VIRT_CPU_ACCOUNTING_GEN 486 bool "Full dynticks CPU time accounting" 487 depends on HAVE_CONTEXT_TRACKING 488 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN 489 depends on GENERIC_CLOCKEVENTS 490 select VIRT_CPU_ACCOUNTING 491 select CONTEXT_TRACKING 492 help 493 Select this option to enable task and CPU time accounting on full 494 dynticks systems. This accounting is implemented by watching every 495 kernel-user boundaries using the context tracking subsystem. 496 The accounting is thus performed at the expense of some significant 497 overhead. 498 499 For now this is only useful if you are working on the full 500 dynticks subsystem development. 501 502 If unsure, say N. 503 504endchoice 505 506config IRQ_TIME_ACCOUNTING 507 bool "Fine granularity task level IRQ time accounting" 508 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE 509 help 510 Select this option to enable fine granularity task irq time 511 accounting. This is done by reading a timestamp on each 512 transitions between softirq and hardirq state, so there can be a 513 small performance impact. 514 515 If in doubt, say N here. 516 517config HAVE_SCHED_AVG_IRQ 518 def_bool y 519 depends on IRQ_TIME_ACCOUNTING || PARAVIRT_TIME_ACCOUNTING 520 depends on SMP 521 522config SCHED_THERMAL_PRESSURE 523 bool 524 default y if ARM && ARM_CPU_TOPOLOGY 525 default y if ARM64 526 depends on SMP 527 depends on CPU_FREQ_THERMAL 528 help 529 Select this option to enable thermal pressure accounting in the 530 scheduler. Thermal pressure is the value conveyed to the scheduler 531 that reflects the reduction in CPU compute capacity resulted from 532 thermal throttling. Thermal throttling occurs when the performance of 533 a CPU is capped due to high operating temperatures. 534 535 If selected, the scheduler will be able to balance tasks accordingly, 536 i.e. put less load on throttled CPUs than on non/less throttled ones. 537 538 This requires the architecture to implement 539 arch_set_thermal_pressure() and arch_scale_thermal_pressure(). 540 541config BSD_PROCESS_ACCT 542 bool "BSD Process Accounting" 543 depends on MULTIUSER 544 help 545 If you say Y here, a user level program will be able to instruct the 546 kernel (via a special system call) to write process accounting 547 information to a file: whenever a process exits, information about 548 that process will be appended to the file by the kernel. The 549 information includes things such as creation time, owning user, 550 command name, memory usage, controlling terminal etc. (the complete 551 list is in the struct acct in <file:include/linux/acct.h>). It is 552 up to the user level program to do useful things with this 553 information. This is generally a good idea, so say Y. 554 555config BSD_PROCESS_ACCT_V3 556 bool "BSD Process Accounting version 3 file format" 557 depends on BSD_PROCESS_ACCT 558 default n 559 help 560 If you say Y here, the process accounting information is written 561 in a new file format that also logs the process IDs of each 562 process and its parent. Note that this file format is incompatible 563 with previous v0/v1/v2 file formats, so you will need updated tools 564 for processing it. A preliminary version of these tools is available 565 at <http://www.gnu.org/software/acct/>. 566 567config TASKSTATS 568 bool "Export task/process statistics through netlink" 569 depends on NET 570 depends on MULTIUSER 571 default n 572 help 573 Export selected statistics for tasks/processes through the 574 generic netlink interface. Unlike BSD process accounting, the 575 statistics are available during the lifetime of tasks/processes as 576 responses to commands. Like BSD accounting, they are sent to user 577 space on task exit. 578 579 Say N if unsure. 580 581config TASK_DELAY_ACCT 582 bool "Enable per-task delay accounting" 583 depends on TASKSTATS 584 select SCHED_INFO 585 help 586 Collect information on time spent by a task waiting for system 587 resources like cpu, synchronous block I/O completion and swapping 588 in pages. Such statistics can help in setting a task's priorities 589 relative to other tasks for cpu, io, rss limits etc. 590 591 Say N if unsure. 592 593config TASK_XACCT 594 bool "Enable extended accounting over taskstats" 595 depends on TASKSTATS 596 help 597 Collect extended task accounting data and send the data 598 to userland for processing over the taskstats interface. 599 600 Say N if unsure. 601 602config TASK_IO_ACCOUNTING 603 bool "Enable per-task storage I/O accounting" 604 depends on TASK_XACCT 605 help 606 Collect information on the number of bytes of storage I/O which this 607 task has caused. 608 609 Say N if unsure. 610 611config PSI 612 bool "Pressure stall information tracking" 613 help 614 Collect metrics that indicate how overcommitted the CPU, memory, 615 and IO capacity are in the system. 616 617 If you say Y here, the kernel will create /proc/pressure/ with the 618 pressure statistics files cpu, memory, and io. These will indicate 619 the share of walltime in which some or all tasks in the system are 620 delayed due to contention of the respective resource. 621 622 In kernels with cgroup support, cgroups (cgroup2 only) will 623 have cpu.pressure, memory.pressure, and io.pressure files, 624 which aggregate pressure stalls for the grouped tasks only. 625 626 For more details see Documentation/accounting/psi.rst. 627 628 Say N if unsure. 629 630config PSI_DEFAULT_DISABLED 631 bool "Require boot parameter to enable pressure stall information tracking" 632 default n 633 depends on PSI 634 help 635 If set, pressure stall information tracking will be disabled 636 per default but can be enabled through passing psi=1 on the 637 kernel commandline during boot. 638 639 This feature adds some code to the task wakeup and sleep 640 paths of the scheduler. The overhead is too low to affect 641 common scheduling-intense workloads in practice (such as 642 webservers, memcache), but it does show up in artificial 643 scheduler stress tests, such as hackbench. 644 645 If you are paranoid and not sure what the kernel will be 646 used for, say Y. 647 648 Say N if unsure. 649 650endmenu # "CPU/Task time and stats accounting" 651 652config CPU_ISOLATION 653 bool "CPU isolation" 654 depends on SMP || COMPILE_TEST 655 default y 656 help 657 Make sure that CPUs running critical tasks are not disturbed by 658 any source of "noise" such as unbound workqueues, timers, kthreads... 659 Unbound jobs get offloaded to housekeeping CPUs. This is driven by 660 the "isolcpus=" boot parameter. 661 662 Say Y if unsure. 663 664source "kernel/rcu/Kconfig" 665 666config BUILD_BIN2C 667 bool 668 default n 669 670config IKCONFIG 671 tristate "Kernel .config support" 672 help 673 This option enables the complete Linux kernel ".config" file 674 contents to be saved in the kernel. It provides documentation 675 of which kernel options are used in a running kernel or in an 676 on-disk kernel. This information can be extracted from the kernel 677 image file with the script scripts/extract-ikconfig and used as 678 input to rebuild the current kernel or to build another kernel. 679 It can also be extracted from a running kernel by reading 680 /proc/config.gz if enabled (below). 681 682config IKCONFIG_PROC 683 bool "Enable access to .config through /proc/config.gz" 684 depends on IKCONFIG && PROC_FS 685 help 686 This option enables access to the kernel configuration file 687 through /proc/config.gz. 688 689config IKHEADERS 690 tristate "Enable kernel headers through /sys/kernel/kheaders.tar.xz" 691 depends on SYSFS 692 help 693 This option enables access to the in-kernel headers that are generated during 694 the build process. These can be used to build eBPF tracing programs, 695 or similar programs. If you build the headers as a module, a module called 696 kheaders.ko is built which can be loaded on-demand to get access to headers. 697 698config LOG_BUF_SHIFT 699 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 700 range 12 25 if !H8300 701 range 12 19 if H8300 702 default 17 703 depends on PRINTK 704 help 705 Select the minimal kernel log buffer size as a power of 2. 706 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config 707 parameter, see below. Any higher size also might be forced 708 by "log_buf_len" boot parameter. 709 710 Examples: 711 17 => 128 KB 712 16 => 64 KB 713 15 => 32 KB 714 14 => 16 KB 715 13 => 8 KB 716 12 => 4 KB 717 718config LOG_CPU_MAX_BUF_SHIFT 719 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)" 720 depends on SMP 721 range 0 21 722 default 12 if !BASE_SMALL 723 default 0 if BASE_SMALL 724 depends on PRINTK 725 help 726 This option allows to increase the default ring buffer size 727 according to the number of CPUs. The value defines the contribution 728 of each CPU as a power of 2. The used space is typically only few 729 lines however it might be much more when problems are reported, 730 e.g. backtraces. 731 732 The increased size means that a new buffer has to be allocated and 733 the original static one is unused. It makes sense only on systems 734 with more CPUs. Therefore this value is used only when the sum of 735 contributions is greater than the half of the default kernel ring 736 buffer as defined by LOG_BUF_SHIFT. The default values are set 737 so that more than 16 CPUs are needed to trigger the allocation. 738 739 Also this option is ignored when "log_buf_len" kernel parameter is 740 used as it forces an exact (power of two) size of the ring buffer. 741 742 The number of possible CPUs is used for this computation ignoring 743 hotplugging making the computation optimal for the worst case 744 scenario while allowing a simple algorithm to be used from bootup. 745 746 Examples shift values and their meaning: 747 17 => 128 KB for each CPU 748 16 => 64 KB for each CPU 749 15 => 32 KB for each CPU 750 14 => 16 KB for each CPU 751 13 => 8 KB for each CPU 752 12 => 4 KB for each CPU 753 754config PRINTK_SAFE_LOG_BUF_SHIFT 755 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)" 756 range 10 21 757 default 13 758 depends on PRINTK 759 help 760 Select the size of an alternate printk per-CPU buffer where messages 761 printed from usafe contexts are temporary stored. One example would 762 be NMI messages, another one - printk recursion. The messages are 763 copied to the main log buffer in a safe context to avoid a deadlock. 764 The value defines the size as a power of 2. 765 766 Those messages are rare and limited. The largest one is when 767 a backtrace is printed. It usually fits into 4KB. Select 768 8KB if you want to be on the safe side. 769 770 Examples: 771 17 => 128 KB for each CPU 772 16 => 64 KB for each CPU 773 15 => 32 KB for each CPU 774 14 => 16 KB for each CPU 775 13 => 8 KB for each CPU 776 12 => 4 KB for each CPU 777 778# 779# Architectures with an unreliable sched_clock() should select this: 780# 781config HAVE_UNSTABLE_SCHED_CLOCK 782 bool 783 784config GENERIC_SCHED_CLOCK 785 bool 786 787menu "Scheduler features" 788 789config UCLAMP_TASK 790 bool "Enable utilization clamping for RT/FAIR tasks" 791 depends on CPU_FREQ_GOV_SCHEDUTIL 792 help 793 This feature enables the scheduler to track the clamped utilization 794 of each CPU based on RUNNABLE tasks scheduled on that CPU. 795 796 With this option, the user can specify the min and max CPU 797 utilization allowed for RUNNABLE tasks. The max utilization defines 798 the maximum frequency a task should use while the min utilization 799 defines the minimum frequency it should use. 800 801 Both min and max utilization clamp values are hints to the scheduler, 802 aiming at improving its frequency selection policy, but they do not 803 enforce or grant any specific bandwidth for tasks. 804 805 If in doubt, say N. 806 807config UCLAMP_BUCKETS_COUNT 808 int "Number of supported utilization clamp buckets" 809 range 5 20 810 default 5 811 depends on UCLAMP_TASK 812 help 813 Defines the number of clamp buckets to use. The range of each bucket 814 will be SCHED_CAPACITY_SCALE/UCLAMP_BUCKETS_COUNT. The higher the 815 number of clamp buckets the finer their granularity and the higher 816 the precision of clamping aggregation and tracking at run-time. 817 818 For example, with the minimum configuration value we will have 5 819 clamp buckets tracking 20% utilization each. A 25% boosted tasks will 820 be refcounted in the [20..39]% bucket and will set the bucket clamp 821 effective value to 25%. 822 If a second 30% boosted task should be co-scheduled on the same CPU, 823 that task will be refcounted in the same bucket of the first task and 824 it will boost the bucket clamp effective value to 30%. 825 The clamp effective value of a bucket is reset to its nominal value 826 (20% in the example above) when there are no more tasks refcounted in 827 that bucket. 828 829 An additional boost/capping margin can be added to some tasks. In the 830 example above the 25% task will be boosted to 30% until it exits the 831 CPU. If that should be considered not acceptable on certain systems, 832 it's always possible to reduce the margin by increasing the number of 833 clamp buckets to trade off used memory for run-time tracking 834 precision. 835 836 If in doubt, use the default value. 837 838endmenu 839 840# 841# For architectures that want to enable the support for NUMA-affine scheduler 842# balancing logic: 843# 844config ARCH_SUPPORTS_NUMA_BALANCING 845 bool 846 847# 848# For architectures that prefer to flush all TLBs after a number of pages 849# are unmapped instead of sending one IPI per page to flush. The architecture 850# must provide guarantees on what happens if a clean TLB cache entry is 851# written after the unmap. Details are in mm/rmap.c near the check for 852# should_defer_flush. The architecture should also consider if the full flush 853# and the refill costs are offset by the savings of sending fewer IPIs. 854config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH 855 bool 856 857config CC_HAS_INT128 858 def_bool !$(cc-option,$(m64-flag) -D__SIZEOF_INT128__=0) && 64BIT 859 860# 861# For architectures that know their GCC __int128 support is sound 862# 863config ARCH_SUPPORTS_INT128 864 bool 865 866# For architectures that (ab)use NUMA to represent different memory regions 867# all cpu-local but of different latencies, such as SuperH. 868# 869config ARCH_WANT_NUMA_VARIABLE_LOCALITY 870 bool 871 872config NUMA_BALANCING 873 bool "Memory placement aware NUMA scheduler" 874 depends on ARCH_SUPPORTS_NUMA_BALANCING 875 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY 876 depends on SMP && NUMA && MIGRATION 877 help 878 This option adds support for automatic NUMA aware memory/task placement. 879 The mechanism is quite primitive and is based on migrating memory when 880 it has references to the node the task is running on. 881 882 This system will be inactive on UMA systems. 883 884config NUMA_BALANCING_DEFAULT_ENABLED 885 bool "Automatically enable NUMA aware memory/task placement" 886 default y 887 depends on NUMA_BALANCING 888 help 889 If set, automatic NUMA balancing will be enabled if running on a NUMA 890 machine. 891 892menuconfig CGROUPS 893 bool "Control Group support" 894 select KERNFS 895 help 896 This option adds support for grouping sets of processes together, for 897 use with process control subsystems such as Cpusets, CFS, memory 898 controls or device isolation. 899 See 900 - Documentation/scheduler/sched-design-CFS.rst (CFS) 901 - Documentation/admin-guide/cgroup-v1/ (features for grouping, isolation 902 and resource control) 903 904 Say N if unsure. 905 906if CGROUPS 907 908config PAGE_COUNTER 909 bool 910 911config MEMCG 912 bool "Memory controller" 913 select PAGE_COUNTER 914 select EVENTFD 915 help 916 Provides control over the memory footprint of tasks in a cgroup. 917 918config MEMCG_SWAP 919 bool 920 depends on MEMCG && SWAP 921 default y 922 923config MEMCG_KMEM 924 bool 925 depends on MEMCG && !SLOB 926 default y 927 928config BLK_CGROUP 929 bool "IO controller" 930 depends on BLOCK 931 default n 932 help 933 Generic block IO controller cgroup interface. This is the common 934 cgroup interface which should be used by various IO controlling 935 policies. 936 937 Currently, CFQ IO scheduler uses it to recognize task groups and 938 control disk bandwidth allocation (proportional time slice allocation) 939 to such task groups. It is also used by bio throttling logic in 940 block layer to implement upper limit in IO rates on a device. 941 942 This option only enables generic Block IO controller infrastructure. 943 One needs to also enable actual IO controlling logic/policy. For 944 enabling proportional weight division of disk bandwidth in CFQ, set 945 CONFIG_BFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 946 CONFIG_BLK_DEV_THROTTLING=y. 947 948 See Documentation/admin-guide/cgroup-v1/blkio-controller.rst for more information. 949 950config CGROUP_WRITEBACK 951 bool 952 depends on MEMCG && BLK_CGROUP 953 default y 954 955menuconfig CGROUP_SCHED 956 bool "CPU controller" 957 default n 958 help 959 This feature lets CPU scheduler recognize task groups and control CPU 960 bandwidth allocation to such task groups. It uses cgroups to group 961 tasks. 962 963if CGROUP_SCHED 964config FAIR_GROUP_SCHED 965 bool "Group scheduling for SCHED_OTHER" 966 depends on CGROUP_SCHED 967 default CGROUP_SCHED 968 969config CFS_BANDWIDTH 970 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 971 depends on FAIR_GROUP_SCHED 972 default n 973 help 974 This option allows users to define CPU bandwidth rates (limits) for 975 tasks running within the fair group scheduler. Groups with no limit 976 set are considered to be unconstrained and will run with no 977 restriction. 978 See Documentation/scheduler/sched-bwc.rst for more information. 979 980config RT_GROUP_SCHED 981 bool "Group scheduling for SCHED_RR/FIFO" 982 depends on CGROUP_SCHED 983 default n 984 help 985 This feature lets you explicitly allocate real CPU bandwidth 986 to task groups. If enabled, it will also make it impossible to 987 schedule realtime tasks for non-root users until you allocate 988 realtime bandwidth for them. 989 See Documentation/scheduler/sched-rt-group.rst for more information. 990 991endif #CGROUP_SCHED 992 993config UCLAMP_TASK_GROUP 994 bool "Utilization clamping per group of tasks" 995 depends on CGROUP_SCHED 996 depends on UCLAMP_TASK 997 default n 998 help 999 This feature enables the scheduler to track the clamped utilization 1000 of each CPU based on RUNNABLE tasks currently scheduled on that CPU. 1001 1002 When this option is enabled, the user can specify a min and max 1003 CPU bandwidth which is allowed for each single task in a group. 1004 The max bandwidth allows to clamp the maximum frequency a task 1005 can use, while the min bandwidth allows to define a minimum 1006 frequency a task will always use. 1007 1008 When task group based utilization clamping is enabled, an eventually 1009 specified task-specific clamp value is constrained by the cgroup 1010 specified clamp value. Both minimum and maximum task clamping cannot 1011 be bigger than the corresponding clamping defined at task group level. 1012 1013 If in doubt, say N. 1014 1015config CGROUP_PIDS 1016 bool "PIDs controller" 1017 help 1018 Provides enforcement of process number limits in the scope of a 1019 cgroup. Any attempt to fork more processes than is allowed in the 1020 cgroup will fail. PIDs are fundamentally a global resource because it 1021 is fairly trivial to reach PID exhaustion before you reach even a 1022 conservative kmemcg limit. As a result, it is possible to grind a 1023 system to halt without being limited by other cgroup policies. The 1024 PIDs controller is designed to stop this from happening. 1025 1026 It should be noted that organisational operations (such as attaching 1027 to a cgroup hierarchy) will *not* be blocked by the PIDs controller, 1028 since the PIDs limit only affects a process's ability to fork, not to 1029 attach to a cgroup. 1030 1031config CGROUP_RDMA 1032 bool "RDMA controller" 1033 help 1034 Provides enforcement of RDMA resources defined by IB stack. 1035 It is fairly easy for consumers to exhaust RDMA resources, which 1036 can result into resource unavailability to other consumers. 1037 RDMA controller is designed to stop this from happening. 1038 Attaching processes with active RDMA resources to the cgroup 1039 hierarchy is allowed even if can cross the hierarchy's limit. 1040 1041config CGROUP_FREEZER 1042 bool "Freezer controller" 1043 help 1044 Provides a way to freeze and unfreeze all tasks in a 1045 cgroup. 1046 1047 This option affects the ORIGINAL cgroup interface. The cgroup2 memory 1048 controller includes important in-kernel memory consumers per default. 1049 1050 If you're using cgroup2, say N. 1051 1052config CGROUP_HUGETLB 1053 bool "HugeTLB controller" 1054 depends on HUGETLB_PAGE 1055 select PAGE_COUNTER 1056 default n 1057 help 1058 Provides a cgroup controller for HugeTLB pages. 1059 When you enable this, you can put a per cgroup limit on HugeTLB usage. 1060 The limit is enforced during page fault. Since HugeTLB doesn't 1061 support page reclaim, enforcing the limit at page fault time implies 1062 that, the application will get SIGBUS signal if it tries to access 1063 HugeTLB pages beyond its limit. This requires the application to know 1064 beforehand how much HugeTLB pages it would require for its use. The 1065 control group is tracked in the third page lru pointer. This means 1066 that we cannot use the controller with huge page less than 3 pages. 1067 1068config CPUSETS 1069 bool "Cpuset controller" 1070 depends on SMP 1071 help 1072 This option will let you create and manage CPUSETs which 1073 allow dynamically partitioning a system into sets of CPUs and 1074 Memory Nodes and assigning tasks to run only within those sets. 1075 This is primarily useful on large SMP or NUMA systems. 1076 1077 Say N if unsure. 1078 1079config PROC_PID_CPUSET 1080 bool "Include legacy /proc/<pid>/cpuset file" 1081 depends on CPUSETS 1082 default y 1083 1084config CGROUP_DEVICE 1085 bool "Device controller" 1086 help 1087 Provides a cgroup controller implementing whitelists for 1088 devices which a process in the cgroup can mknod or open. 1089 1090config CGROUP_CPUACCT 1091 bool "Simple CPU accounting controller" 1092 help 1093 Provides a simple controller for monitoring the 1094 total CPU consumed by the tasks in a cgroup. 1095 1096config CGROUP_PERF 1097 bool "Perf controller" 1098 depends on PERF_EVENTS 1099 help 1100 This option extends the perf per-cpu mode to restrict monitoring 1101 to threads which belong to the cgroup specified and run on the 1102 designated cpu. Or this can be used to have cgroup ID in samples 1103 so that it can monitor performance events among cgroups. 1104 1105 Say N if unsure. 1106 1107config CGROUP_BPF 1108 bool "Support for eBPF programs attached to cgroups" 1109 depends on BPF_SYSCALL 1110 select SOCK_CGROUP_DATA 1111 help 1112 Allow attaching eBPF programs to a cgroup using the bpf(2) 1113 syscall command BPF_PROG_ATTACH. 1114 1115 In which context these programs are accessed depends on the type 1116 of attachment. For instance, programs that are attached using 1117 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of 1118 inet sockets. 1119 1120config CGROUP_MISC 1121 bool "Misc resource controller" 1122 default n 1123 help 1124 Provides a controller for miscellaneous resources on a host. 1125 1126 Miscellaneous scalar resources are the resources on the host system 1127 which cannot be abstracted like the other cgroups. This controller 1128 tracks and limits the miscellaneous resources used by a process 1129 attached to a cgroup hierarchy. 1130 1131 For more information, please check misc cgroup section in 1132 /Documentation/admin-guide/cgroup-v2.rst. 1133 1134config CGROUP_DEBUG 1135 bool "Debug controller" 1136 default n 1137 depends on DEBUG_KERNEL 1138 help 1139 This option enables a simple controller that exports 1140 debugging information about the cgroups framework. This 1141 controller is for control cgroup debugging only. Its 1142 interfaces are not stable. 1143 1144 Say N. 1145 1146config SOCK_CGROUP_DATA 1147 bool 1148 default n 1149 1150endif # CGROUPS 1151 1152menuconfig NAMESPACES 1153 bool "Namespaces support" if EXPERT 1154 depends on MULTIUSER 1155 default !EXPERT 1156 help 1157 Provides the way to make tasks work with different objects using 1158 the same id. For example same IPC id may refer to different objects 1159 or same user id or pid may refer to different tasks when used in 1160 different namespaces. 1161 1162if NAMESPACES 1163 1164config UTS_NS 1165 bool "UTS namespace" 1166 default y 1167 help 1168 In this namespace tasks see different info provided with the 1169 uname() system call 1170 1171config TIME_NS 1172 bool "TIME namespace" 1173 depends on GENERIC_VDSO_TIME_NS 1174 default y 1175 help 1176 In this namespace boottime and monotonic clocks can be set. 1177 The time will keep going with the same pace. 1178 1179config IPC_NS 1180 bool "IPC namespace" 1181 depends on (SYSVIPC || POSIX_MQUEUE) 1182 default y 1183 help 1184 In this namespace tasks work with IPC ids which correspond to 1185 different IPC objects in different namespaces. 1186 1187config USER_NS 1188 bool "User namespace" 1189 default n 1190 help 1191 This allows containers, i.e. vservers, to use user namespaces 1192 to provide different user info for different servers. 1193 1194 When user namespaces are enabled in the kernel it is 1195 recommended that the MEMCG option also be enabled and that 1196 user-space use the memory control groups to limit the amount 1197 of memory a memory unprivileged users can use. 1198 1199 If unsure, say N. 1200 1201config PID_NS 1202 bool "PID Namespaces" 1203 default y 1204 help 1205 Support process id namespaces. This allows having multiple 1206 processes with the same pid as long as they are in different 1207 pid namespaces. This is a building block of containers. 1208 1209config NET_NS 1210 bool "Network namespace" 1211 depends on NET 1212 default y 1213 help 1214 Allow user space to create what appear to be multiple instances 1215 of the network stack. 1216 1217endif # NAMESPACES 1218 1219config CHECKPOINT_RESTORE 1220 bool "Checkpoint/restore support" 1221 select PROC_CHILDREN 1222 select KCMP 1223 default n 1224 help 1225 Enables additional kernel features in a sake of checkpoint/restore. 1226 In particular it adds auxiliary prctl codes to setup process text, 1227 data and heap segment sizes, and a few additional /proc filesystem 1228 entries. 1229 1230 If unsure, say N here. 1231 1232config SCHED_AUTOGROUP 1233 bool "Automatic process group scheduling" 1234 select CGROUPS 1235 select CGROUP_SCHED 1236 select FAIR_GROUP_SCHED 1237 help 1238 This option optimizes the scheduler for common desktop workloads by 1239 automatically creating and populating task groups. This separation 1240 of workloads isolates aggressive CPU burners (like build jobs) from 1241 desktop applications. Task group autogeneration is currently based 1242 upon task session. 1243 1244config SYSFS_DEPRECATED 1245 bool "Enable deprecated sysfs features to support old userspace tools" 1246 depends on SYSFS 1247 default n 1248 help 1249 This option adds code that switches the layout of the "block" class 1250 devices, to not show up in /sys/class/block/, but only in 1251 /sys/block/. 1252 1253 This switch is only active when the sysfs.deprecated=1 boot option is 1254 passed or the SYSFS_DEPRECATED_V2 option is set. 1255 1256 This option allows new kernels to run on old distributions and tools, 1257 which might get confused by /sys/class/block/. Since 2007/2008 all 1258 major distributions and tools handle this just fine. 1259 1260 Recent distributions and userspace tools after 2009/2010 depend on 1261 the existence of /sys/class/block/, and will not work with this 1262 option enabled. 1263 1264 Only if you are using a new kernel on an old distribution, you might 1265 need to say Y here. 1266 1267config SYSFS_DEPRECATED_V2 1268 bool "Enable deprecated sysfs features by default" 1269 default n 1270 depends on SYSFS 1271 depends on SYSFS_DEPRECATED 1272 help 1273 Enable deprecated sysfs by default. 1274 1275 See the CONFIG_SYSFS_DEPRECATED option for more details about this 1276 option. 1277 1278 Only if you are using a new kernel on an old distribution, you might 1279 need to say Y here. Even then, odds are you would not need it 1280 enabled, you can always pass the boot option if absolutely necessary. 1281 1282config RELAY 1283 bool "Kernel->user space relay support (formerly relayfs)" 1284 select IRQ_WORK 1285 help 1286 This option enables support for relay interface support in 1287 certain file systems (such as debugfs). 1288 It is designed to provide an efficient mechanism for tools and 1289 facilities to relay large amounts of data from kernel space to 1290 user space. 1291 1292 If unsure, say N. 1293 1294config BLK_DEV_INITRD 1295 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1296 help 1297 The initial RAM filesystem is a ramfs which is loaded by the 1298 boot loader (loadlin or lilo) and that is mounted as root 1299 before the normal boot procedure. It is typically used to 1300 load modules needed to mount the "real" root file system, 1301 etc. See <file:Documentation/admin-guide/initrd.rst> for details. 1302 1303 If RAM disk support (BLK_DEV_RAM) is also included, this 1304 also enables initial RAM disk (initrd) support and adds 1305 15 Kbytes (more on some other architectures) to the kernel size. 1306 1307 If unsure say Y. 1308 1309if BLK_DEV_INITRD 1310 1311source "usr/Kconfig" 1312 1313endif 1314 1315config BOOT_CONFIG 1316 bool "Boot config support" 1317 select BLK_DEV_INITRD 1318 help 1319 Extra boot config allows system admin to pass a config file as 1320 complemental extension of kernel cmdline when booting. 1321 The boot config file must be attached at the end of initramfs 1322 with checksum, size and magic word. 1323 See <file:Documentation/admin-guide/bootconfig.rst> for details. 1324 1325 If unsure, say Y. 1326 1327choice 1328 prompt "Compiler optimization level" 1329 default CC_OPTIMIZE_FOR_PERFORMANCE 1330 1331config CC_OPTIMIZE_FOR_PERFORMANCE 1332 bool "Optimize for performance (-O2)" 1333 help 1334 This is the default optimization level for the kernel, building 1335 with the "-O2" compiler flag for best performance and most 1336 helpful compile-time warnings. 1337 1338config CC_OPTIMIZE_FOR_PERFORMANCE_O3 1339 bool "Optimize more for performance (-O3)" 1340 depends on ARC 1341 help 1342 Choosing this option will pass "-O3" to your compiler to optimize 1343 the kernel yet more for performance. 1344 1345config CC_OPTIMIZE_FOR_SIZE 1346 bool "Optimize for size (-Os)" 1347 help 1348 Choosing this option will pass "-Os" to your compiler resulting 1349 in a smaller kernel. 1350 1351endchoice 1352 1353config HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1354 bool 1355 help 1356 This requires that the arch annotates or otherwise protects 1357 its external entry points from being discarded. Linker scripts 1358 must also merge .text.*, .data.*, and .bss.* correctly into 1359 output sections. Care must be taken not to pull in unrelated 1360 sections (e.g., '.text.init'). Typically '.' in section names 1361 is used to distinguish them from label names / C identifiers. 1362 1363config LD_DEAD_CODE_DATA_ELIMINATION 1364 bool "Dead code and data elimination (EXPERIMENTAL)" 1365 depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1366 depends on EXPERT 1367 depends on $(cc-option,-ffunction-sections -fdata-sections) 1368 depends on $(ld-option,--gc-sections) 1369 help 1370 Enable this if you want to do dead code and data elimination with 1371 the linker by compiling with -ffunction-sections -fdata-sections, 1372 and linking with --gc-sections. 1373 1374 This can reduce on disk and in-memory size of the kernel 1375 code and static data, particularly for small configs and 1376 on small systems. This has the possibility of introducing 1377 silently broken kernel if the required annotations are not 1378 present. This option is not well tested yet, so use at your 1379 own risk. 1380 1381config LD_ORPHAN_WARN 1382 def_bool y 1383 depends on ARCH_WANT_LD_ORPHAN_WARN 1384 depends on !LD_IS_LLD || LLD_VERSION >= 110000 1385 depends on $(ld-option,--orphan-handling=warn) 1386 1387config SYSCTL 1388 bool 1389 1390config HAVE_UID16 1391 bool 1392 1393config SYSCTL_EXCEPTION_TRACE 1394 bool 1395 help 1396 Enable support for /proc/sys/debug/exception-trace. 1397 1398config SYSCTL_ARCH_UNALIGN_NO_WARN 1399 bool 1400 help 1401 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap 1402 Allows arch to define/use @no_unaligned_warning to possibly warn 1403 about unaligned access emulation going on under the hood. 1404 1405config SYSCTL_ARCH_UNALIGN_ALLOW 1406 bool 1407 help 1408 Enable support for /proc/sys/kernel/unaligned-trap 1409 Allows arches to define/use @unaligned_enabled to runtime toggle 1410 the unaligned access emulation. 1411 see arch/parisc/kernel/unaligned.c for reference 1412 1413config HAVE_PCSPKR_PLATFORM 1414 bool 1415 1416# interpreter that classic socket filters depend on 1417config BPF 1418 bool 1419 1420menuconfig EXPERT 1421 bool "Configure standard kernel features (expert users)" 1422 # Unhide debug options, to make the on-by-default options visible 1423 select DEBUG_KERNEL 1424 help 1425 This option allows certain base kernel options and settings 1426 to be disabled or tweaked. This is for specialized 1427 environments which can tolerate a "non-standard" kernel. 1428 Only use this if you really know what you are doing. 1429 1430config UID16 1431 bool "Enable 16-bit UID system calls" if EXPERT 1432 depends on HAVE_UID16 && MULTIUSER 1433 default y 1434 help 1435 This enables the legacy 16-bit UID syscall wrappers. 1436 1437config MULTIUSER 1438 bool "Multiple users, groups and capabilities support" if EXPERT 1439 default y 1440 help 1441 This option enables support for non-root users, groups and 1442 capabilities. 1443 1444 If you say N here, all processes will run with UID 0, GID 0, and all 1445 possible capabilities. Saying N here also compiles out support for 1446 system calls related to UIDs, GIDs, and capabilities, such as setuid, 1447 setgid, and capset. 1448 1449 If unsure, say Y here. 1450 1451config SGETMASK_SYSCALL 1452 bool "sgetmask/ssetmask syscalls support" if EXPERT 1453 def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH 1454 help 1455 sys_sgetmask and sys_ssetmask are obsolete system calls 1456 no longer supported in libc but still enabled by default in some 1457 architectures. 1458 1459 If unsure, leave the default option here. 1460 1461config SYSFS_SYSCALL 1462 bool "Sysfs syscall support" if EXPERT 1463 default y 1464 help 1465 sys_sysfs is an obsolete system call no longer supported in libc. 1466 Note that disabling this option is more secure but might break 1467 compatibility with some systems. 1468 1469 If unsure say Y here. 1470 1471config FHANDLE 1472 bool "open by fhandle syscalls" if EXPERT 1473 select EXPORTFS 1474 default y 1475 help 1476 If you say Y here, a user level program will be able to map 1477 file names to handle and then later use the handle for 1478 different file system operations. This is useful in implementing 1479 userspace file servers, which now track files using handles instead 1480 of names. The handle would remain the same even if file names 1481 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 1482 syscalls. 1483 1484config POSIX_TIMERS 1485 bool "Posix Clocks & timers" if EXPERT 1486 default y 1487 help 1488 This includes native support for POSIX timers to the kernel. 1489 Some embedded systems have no use for them and therefore they 1490 can be configured out to reduce the size of the kernel image. 1491 1492 When this option is disabled, the following syscalls won't be 1493 available: timer_create, timer_gettime: timer_getoverrun, 1494 timer_settime, timer_delete, clock_adjtime, getitimer, 1495 setitimer, alarm. Furthermore, the clock_settime, clock_gettime, 1496 clock_getres and clock_nanosleep syscalls will be limited to 1497 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only. 1498 1499 If unsure say y. 1500 1501config PRINTK 1502 default y 1503 bool "Enable support for printk" if EXPERT 1504 select IRQ_WORK 1505 help 1506 This option enables normal printk support. Removing it 1507 eliminates most of the message strings from the kernel image 1508 and makes the kernel more or less silent. As this makes it 1509 very difficult to diagnose system problems, saying N here is 1510 strongly discouraged. 1511 1512config PRINTK_NMI 1513 def_bool y 1514 depends on PRINTK 1515 depends on HAVE_NMI 1516 1517config BUG 1518 bool "BUG() support" if EXPERT 1519 default y 1520 help 1521 Disabling this option eliminates support for BUG and WARN, reducing 1522 the size of your kernel image and potentially quietly ignoring 1523 numerous fatal conditions. You should only consider disabling this 1524 option for embedded systems with no facilities for reporting errors. 1525 Just say Y. 1526 1527config ELF_CORE 1528 depends on COREDUMP 1529 default y 1530 bool "Enable ELF core dumps" if EXPERT 1531 help 1532 Enable support for generating core dumps. Disabling saves about 4k. 1533 1534 1535config PCSPKR_PLATFORM 1536 bool "Enable PC-Speaker support" if EXPERT 1537 depends on HAVE_PCSPKR_PLATFORM 1538 select I8253_LOCK 1539 default y 1540 help 1541 This option allows to disable the internal PC-Speaker 1542 support, saving some memory. 1543 1544config BASE_FULL 1545 default y 1546 bool "Enable full-sized data structures for core" if EXPERT 1547 help 1548 Disabling this option reduces the size of miscellaneous core 1549 kernel data structures. This saves memory on small machines, 1550 but may reduce performance. 1551 1552config FUTEX 1553 bool "Enable futex support" if EXPERT 1554 default y 1555 imply RT_MUTEXES 1556 help 1557 Disabling this option will cause the kernel to be built without 1558 support for "fast userspace mutexes". The resulting kernel may not 1559 run glibc-based applications correctly. 1560 1561config FUTEX_PI 1562 bool 1563 depends on FUTEX && RT_MUTEXES 1564 default y 1565 1566config HAVE_FUTEX_CMPXCHG 1567 bool 1568 depends on FUTEX 1569 help 1570 Architectures should select this if futex_atomic_cmpxchg_inatomic() 1571 is implemented and always working. This removes a couple of runtime 1572 checks. 1573 1574config EPOLL 1575 bool "Enable eventpoll support" if EXPERT 1576 default y 1577 help 1578 Disabling this option will cause the kernel to be built without 1579 support for epoll family of system calls. 1580 1581config SIGNALFD 1582 bool "Enable signalfd() system call" if EXPERT 1583 default y 1584 help 1585 Enable the signalfd() system call that allows to receive signals 1586 on a file descriptor. 1587 1588 If unsure, say Y. 1589 1590config TIMERFD 1591 bool "Enable timerfd() system call" if EXPERT 1592 default y 1593 help 1594 Enable the timerfd() system call that allows to receive timer 1595 events on a file descriptor. 1596 1597 If unsure, say Y. 1598 1599config EVENTFD 1600 bool "Enable eventfd() system call" if EXPERT 1601 default y 1602 help 1603 Enable the eventfd() system call that allows to receive both 1604 kernel notification (ie. KAIO) or userspace notifications. 1605 1606 If unsure, say Y. 1607 1608config SHMEM 1609 bool "Use full shmem filesystem" if EXPERT 1610 default y 1611 depends on MMU 1612 help 1613 The shmem is an internal filesystem used to manage shared memory. 1614 It is backed by swap and manages resource limits. It is also exported 1615 to userspace as tmpfs if TMPFS is enabled. Disabling this 1616 option replaces shmem and tmpfs with the much simpler ramfs code, 1617 which may be appropriate on small systems without swap. 1618 1619config AIO 1620 bool "Enable AIO support" if EXPERT 1621 default y 1622 help 1623 This option enables POSIX asynchronous I/O which may by used 1624 by some high performance threaded applications. Disabling 1625 this option saves about 7k. 1626 1627config IO_URING 1628 bool "Enable IO uring support" if EXPERT 1629 select IO_WQ 1630 default y 1631 help 1632 This option enables support for the io_uring interface, enabling 1633 applications to submit and complete IO through submission and 1634 completion rings that are shared between the kernel and application. 1635 1636config ADVISE_SYSCALLS 1637 bool "Enable madvise/fadvise syscalls" if EXPERT 1638 default y 1639 help 1640 This option enables the madvise and fadvise syscalls, used by 1641 applications to advise the kernel about their future memory or file 1642 usage, improving performance. If building an embedded system where no 1643 applications use these syscalls, you can disable this option to save 1644 space. 1645 1646config HAVE_ARCH_USERFAULTFD_WP 1647 bool 1648 help 1649 Arch has userfaultfd write protection support 1650 1651config HAVE_ARCH_USERFAULTFD_MINOR 1652 bool 1653 help 1654 Arch has userfaultfd minor fault support 1655 1656config MEMBARRIER 1657 bool "Enable membarrier() system call" if EXPERT 1658 default y 1659 help 1660 Enable the membarrier() system call that allows issuing memory 1661 barriers across all running threads, which can be used to distribute 1662 the cost of user-space memory barriers asymmetrically by transforming 1663 pairs of memory barriers into pairs consisting of membarrier() and a 1664 compiler barrier. 1665 1666 If unsure, say Y. 1667 1668config KALLSYMS 1669 bool "Load all symbols for debugging/ksymoops" if EXPERT 1670 default y 1671 help 1672 Say Y here to let the kernel print out symbolic crash information and 1673 symbolic stack backtraces. This increases the size of the kernel 1674 somewhat, as all symbols have to be loaded into the kernel image. 1675 1676config KALLSYMS_ALL 1677 bool "Include all symbols in kallsyms" 1678 depends on DEBUG_KERNEL && KALLSYMS 1679 help 1680 Normally kallsyms only contains the symbols of functions for nicer 1681 OOPS messages and backtraces (i.e., symbols from the text and inittext 1682 sections). This is sufficient for most cases. And only in very rare 1683 cases (e.g., when a debugger is used) all symbols are required (e.g., 1684 names of variables from the data sections, etc). 1685 1686 This option makes sure that all symbols are loaded into the kernel 1687 image (i.e., symbols from all sections) in cost of increased kernel 1688 size (depending on the kernel configuration, it may be 300KiB or 1689 something like this). 1690 1691 Say N unless you really need all symbols. 1692 1693config KALLSYMS_ABSOLUTE_PERCPU 1694 bool 1695 depends on KALLSYMS 1696 default X86_64 && SMP 1697 1698config KALLSYMS_BASE_RELATIVE 1699 bool 1700 depends on KALLSYMS 1701 default !IA64 1702 help 1703 Instead of emitting them as absolute values in the native word size, 1704 emit the symbol references in the kallsyms table as 32-bit entries, 1705 each containing a relative value in the range [base, base + U32_MAX] 1706 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either 1707 an absolute value in the range [0, S32_MAX] or a relative value in the 1708 range [base, base + S32_MAX], where base is the lowest relative symbol 1709 address encountered in the image. 1710 1711 On 64-bit builds, this reduces the size of the address table by 50%, 1712 but more importantly, it results in entries whose values are build 1713 time constants, and no relocation pass is required at runtime to fix 1714 up the entries based on the runtime load address of the kernel. 1715 1716# end of the "standard kernel features (expert users)" menu 1717 1718# syscall, maps, verifier 1719 1720config USERFAULTFD 1721 bool "Enable userfaultfd() system call" 1722 depends on MMU 1723 help 1724 Enable the userfaultfd() system call that allows to intercept and 1725 handle page faults in userland. 1726 1727config ARCH_HAS_MEMBARRIER_CALLBACKS 1728 bool 1729 1730config ARCH_HAS_MEMBARRIER_SYNC_CORE 1731 bool 1732 1733config KCMP 1734 bool "Enable kcmp() system call" if EXPERT 1735 help 1736 Enable the kernel resource comparison system call. It provides 1737 user-space with the ability to compare two processes to see if they 1738 share a common resource, such as a file descriptor or even virtual 1739 memory space. 1740 1741 If unsure, say N. 1742 1743config RSEQ 1744 bool "Enable rseq() system call" if EXPERT 1745 default y 1746 depends on HAVE_RSEQ 1747 select MEMBARRIER 1748 help 1749 Enable the restartable sequences system call. It provides a 1750 user-space cache for the current CPU number value, which 1751 speeds up getting the current CPU number from user-space, 1752 as well as an ABI to speed up user-space operations on 1753 per-CPU data. 1754 1755 If unsure, say Y. 1756 1757config DEBUG_RSEQ 1758 default n 1759 bool "Enabled debugging of rseq() system call" if EXPERT 1760 depends on RSEQ && DEBUG_KERNEL 1761 help 1762 Enable extra debugging checks for the rseq system call. 1763 1764 If unsure, say N. 1765 1766config EMBEDDED 1767 bool "Embedded system" 1768 select EXPERT 1769 help 1770 This option should be enabled if compiling the kernel for 1771 an embedded system so certain expert options are available 1772 for configuration. 1773 1774config HAVE_PERF_EVENTS 1775 bool 1776 help 1777 See tools/perf/design.txt for details. 1778 1779config PERF_USE_VMALLOC 1780 bool 1781 help 1782 See tools/perf/design.txt for details 1783 1784config PC104 1785 bool "PC/104 support" if EXPERT 1786 help 1787 Expose PC/104 form factor device drivers and options available for 1788 selection and configuration. Enable this option if your target 1789 machine has a PC/104 bus. 1790 1791menu "Kernel Performance Events And Counters" 1792 1793config PERF_EVENTS 1794 bool "Kernel performance events and counters" 1795 default y if PROFILING 1796 depends on HAVE_PERF_EVENTS 1797 select IRQ_WORK 1798 select SRCU 1799 help 1800 Enable kernel support for various performance events provided 1801 by software and hardware. 1802 1803 Software events are supported either built-in or via the 1804 use of generic tracepoints. 1805 1806 Most modern CPUs support performance events via performance 1807 counter registers. These registers count the number of certain 1808 types of hw events: such as instructions executed, cachemisses 1809 suffered, or branches mis-predicted - without slowing down the 1810 kernel or applications. These registers can also trigger interrupts 1811 when a threshold number of events have passed - and can thus be 1812 used to profile the code that runs on that CPU. 1813 1814 The Linux Performance Event subsystem provides an abstraction of 1815 these software and hardware event capabilities, available via a 1816 system call and used by the "perf" utility in tools/perf/. It 1817 provides per task and per CPU counters, and it provides event 1818 capabilities on top of those. 1819 1820 Say Y if unsure. 1821 1822config DEBUG_PERF_USE_VMALLOC 1823 default n 1824 bool "Debug: use vmalloc to back perf mmap() buffers" 1825 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC 1826 select PERF_USE_VMALLOC 1827 help 1828 Use vmalloc memory to back perf mmap() buffers. 1829 1830 Mostly useful for debugging the vmalloc code on platforms 1831 that don't require it. 1832 1833 Say N if unsure. 1834 1835endmenu 1836 1837config VM_EVENT_COUNTERS 1838 default y 1839 bool "Enable VM event counters for /proc/vmstat" if EXPERT 1840 help 1841 VM event counters are needed for event counts to be shown. 1842 This option allows the disabling of the VM event counters 1843 on EXPERT systems. /proc/vmstat will only show page counts 1844 if VM event counters are disabled. 1845 1846config SLUB_DEBUG 1847 default y 1848 bool "Enable SLUB debugging support" if EXPERT 1849 depends on SLUB && SYSFS 1850 select STACKDEPOT if STACKTRACE_SUPPORT 1851 help 1852 SLUB has extensive debug support features. Disabling these can 1853 result in significant savings in code size. This also disables 1854 SLUB sysfs support. /sys/slab will not exist and there will be 1855 no support for cache validation etc. 1856 1857config COMPAT_BRK 1858 bool "Disable heap randomization" 1859 default y 1860 help 1861 Randomizing heap placement makes heap exploits harder, but it 1862 also breaks ancient binaries (including anything libc5 based). 1863 This option changes the bootup default to heap randomization 1864 disabled, and can be overridden at runtime by setting 1865 /proc/sys/kernel/randomize_va_space to 2. 1866 1867 On non-ancient distros (post-2000 ones) N is usually a safe choice. 1868 1869choice 1870 prompt "Choose SLAB allocator" 1871 default SLUB 1872 help 1873 This option allows to select a slab allocator. 1874 1875config SLAB 1876 bool "SLAB" 1877 select HAVE_HARDENED_USERCOPY_ALLOCATOR 1878 help 1879 The regular slab allocator that is established and known to work 1880 well in all environments. It organizes cache hot objects in 1881 per cpu and per node queues. 1882 1883config SLUB 1884 bool "SLUB (Unqueued Allocator)" 1885 select HAVE_HARDENED_USERCOPY_ALLOCATOR 1886 help 1887 SLUB is a slab allocator that minimizes cache line usage 1888 instead of managing queues of cached objects (SLAB approach). 1889 Per cpu caching is realized using slabs of objects instead 1890 of queues of objects. SLUB can use memory efficiently 1891 and has enhanced diagnostics. SLUB is the default choice for 1892 a slab allocator. 1893 1894config SLOB 1895 depends on EXPERT 1896 bool "SLOB (Simple Allocator)" 1897 help 1898 SLOB replaces the stock allocator with a drastically simpler 1899 allocator. SLOB is generally more space efficient but 1900 does not perform as well on large systems. 1901 1902endchoice 1903 1904config SLAB_MERGE_DEFAULT 1905 bool "Allow slab caches to be merged" 1906 default y 1907 help 1908 For reduced kernel memory fragmentation, slab caches can be 1909 merged when they share the same size and other characteristics. 1910 This carries a risk of kernel heap overflows being able to 1911 overwrite objects from merged caches (and more easily control 1912 cache layout), which makes such heap attacks easier to exploit 1913 by attackers. By keeping caches unmerged, these kinds of exploits 1914 can usually only damage objects in the same cache. To disable 1915 merging at runtime, "slab_nomerge" can be passed on the kernel 1916 command line. 1917 1918config SLAB_FREELIST_RANDOM 1919 bool "Randomize slab freelist" 1920 depends on SLAB || SLUB 1921 help 1922 Randomizes the freelist order used on creating new pages. This 1923 security feature reduces the predictability of the kernel slab 1924 allocator against heap overflows. 1925 1926config SLAB_FREELIST_HARDENED 1927 bool "Harden slab freelist metadata" 1928 depends on SLAB || SLUB 1929 help 1930 Many kernel heap attacks try to target slab cache metadata and 1931 other infrastructure. This options makes minor performance 1932 sacrifices to harden the kernel slab allocator against common 1933 freelist exploit methods. Some slab implementations have more 1934 sanity-checking than others. This option is most effective with 1935 CONFIG_SLUB. 1936 1937config SHUFFLE_PAGE_ALLOCATOR 1938 bool "Page allocator randomization" 1939 default SLAB_FREELIST_RANDOM && ACPI_NUMA 1940 help 1941 Randomization of the page allocator improves the average 1942 utilization of a direct-mapped memory-side-cache. See section 1943 5.2.27 Heterogeneous Memory Attribute Table (HMAT) in the ACPI 1944 6.2a specification for an example of how a platform advertises 1945 the presence of a memory-side-cache. There are also incidental 1946 security benefits as it reduces the predictability of page 1947 allocations to compliment SLAB_FREELIST_RANDOM, but the 1948 default granularity of shuffling on the "MAX_ORDER - 1" i.e, 1949 10th order of pages is selected based on cache utilization 1950 benefits on x86. 1951 1952 While the randomization improves cache utilization it may 1953 negatively impact workloads on platforms without a cache. For 1954 this reason, by default, the randomization is enabled only 1955 after runtime detection of a direct-mapped memory-side-cache. 1956 Otherwise, the randomization may be force enabled with the 1957 'page_alloc.shuffle' kernel command line parameter. 1958 1959 Say Y if unsure. 1960 1961config SLUB_CPU_PARTIAL 1962 default y 1963 depends on SLUB && SMP 1964 bool "SLUB per cpu partial cache" 1965 help 1966 Per cpu partial caches accelerate objects allocation and freeing 1967 that is local to a processor at the price of more indeterminism 1968 in the latency of the free. On overflow these caches will be cleared 1969 which requires the taking of locks that may cause latency spikes. 1970 Typically one would choose no for a realtime system. 1971 1972config MMAP_ALLOW_UNINITIALIZED 1973 bool "Allow mmapped anonymous memory to be uninitialized" 1974 depends on EXPERT && !MMU 1975 default n 1976 help 1977 Normally, and according to the Linux spec, anonymous memory obtained 1978 from mmap() has its contents cleared before it is passed to 1979 userspace. Enabling this config option allows you to request that 1980 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus 1981 providing a huge performance boost. If this option is not enabled, 1982 then the flag will be ignored. 1983 1984 This is taken advantage of by uClibc's malloc(), and also by 1985 ELF-FDPIC binfmt's brk and stack allocator. 1986 1987 Because of the obvious security issues, this option should only be 1988 enabled on embedded devices where you control what is run in 1989 userspace. Since that isn't generally a problem on no-MMU systems, 1990 it is normally safe to say Y here. 1991 1992 See Documentation/admin-guide/mm/nommu-mmap.rst for more information. 1993 1994config SYSTEM_DATA_VERIFICATION 1995 def_bool n 1996 select SYSTEM_TRUSTED_KEYRING 1997 select KEYS 1998 select CRYPTO 1999 select CRYPTO_RSA 2000 select ASYMMETRIC_KEY_TYPE 2001 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE 2002 select ASN1 2003 select OID_REGISTRY 2004 select X509_CERTIFICATE_PARSER 2005 select PKCS7_MESSAGE_PARSER 2006 help 2007 Provide PKCS#7 message verification using the contents of the system 2008 trusted keyring to provide public keys. This then can be used for 2009 module verification, kexec image verification and firmware blob 2010 verification. 2011 2012config PROFILING 2013 bool "Profiling support" 2014 help 2015 Say Y here to enable the extended profiling support mechanisms used 2016 by profilers. 2017 2018# 2019# Place an empty function call at each tracepoint site. Can be 2020# dynamically changed for a probe function. 2021# 2022config TRACEPOINTS 2023 bool 2024 2025endmenu # General setup 2026 2027source "arch/Kconfig" 2028 2029config RT_MUTEXES 2030 bool 2031 2032config BASE_SMALL 2033 int 2034 default 0 if BASE_FULL 2035 default 1 if !BASE_FULL 2036 2037config MODULE_SIG_FORMAT 2038 def_bool n 2039 select SYSTEM_DATA_VERIFICATION 2040 2041menuconfig MODULES 2042 bool "Enable loadable module support" 2043 modules 2044 help 2045 Kernel modules are small pieces of compiled code which can 2046 be inserted in the running kernel, rather than being 2047 permanently built into the kernel. You use the "modprobe" 2048 tool to add (and sometimes remove) them. If you say Y here, 2049 many parts of the kernel can be built as modules (by 2050 answering M instead of Y where indicated): this is most 2051 useful for infrequently used options which are not required 2052 for booting. For more information, see the man pages for 2053 modprobe, lsmod, modinfo, insmod and rmmod. 2054 2055 If you say Y here, you will need to run "make 2056 modules_install" to put the modules under /lib/modules/ 2057 where modprobe can find them (you may need to be root to do 2058 this). 2059 2060 If unsure, say Y. 2061 2062if MODULES 2063 2064config MODULE_FORCE_LOAD 2065 bool "Forced module loading" 2066 default n 2067 help 2068 Allow loading of modules without version information (ie. modprobe 2069 --force). Forced module loading sets the 'F' (forced) taint flag and 2070 is usually a really bad idea. 2071 2072config MODULE_UNLOAD 2073 bool "Module unloading" 2074 help 2075 Without this option you will not be able to unload any 2076 modules (note that some modules may not be unloadable 2077 anyway), which makes your kernel smaller, faster 2078 and simpler. If unsure, say Y. 2079 2080config MODULE_FORCE_UNLOAD 2081 bool "Forced module unloading" 2082 depends on MODULE_UNLOAD 2083 help 2084 This option allows you to force a module to unload, even if the 2085 kernel believes it is unsafe: the kernel will remove the module 2086 without waiting for anyone to stop using it (using the -f option to 2087 rmmod). This is mainly for kernel developers and desperate users. 2088 If unsure, say N. 2089 2090config MODVERSIONS 2091 bool "Module versioning support" 2092 help 2093 Usually, you have to use modules compiled with your kernel. 2094 Saying Y here makes it sometimes possible to use modules 2095 compiled for different kernels, by adding enough information 2096 to the modules to (hopefully) spot any changes which would 2097 make them incompatible with the kernel you are running. If 2098 unsure, say N. 2099 2100config ASM_MODVERSIONS 2101 bool 2102 default HAVE_ASM_MODVERSIONS && MODVERSIONS 2103 help 2104 This enables module versioning for exported symbols also from 2105 assembly. This can be enabled only when the target architecture 2106 supports it. 2107 2108config MODULE_REL_CRCS 2109 bool 2110 depends on MODVERSIONS 2111 2112config MODULE_SRCVERSION_ALL 2113 bool "Source checksum for all modules" 2114 help 2115 Modules which contain a MODULE_VERSION get an extra "srcversion" 2116 field inserted into their modinfo section, which contains a 2117 sum of the source files which made it. This helps maintainers 2118 see exactly which source was used to build a module (since 2119 others sometimes change the module source without updating 2120 the version). With this option, such a "srcversion" field 2121 will be created for all modules. If unsure, say N. 2122 2123config MODULE_SIG 2124 bool "Module signature verification" 2125 select MODULE_SIG_FORMAT 2126 help 2127 Check modules for valid signatures upon load: the signature 2128 is simply appended to the module. For more information see 2129 <file:Documentation/admin-guide/module-signing.rst>. 2130 2131 Note that this option adds the OpenSSL development packages as a 2132 kernel build dependency so that the signing tool can use its crypto 2133 library. 2134 2135 You should enable this option if you wish to use either 2136 CONFIG_SECURITY_LOCKDOWN_LSM or lockdown functionality imposed via 2137 another LSM - otherwise unsigned modules will be loadable regardless 2138 of the lockdown policy. 2139 2140 !!!WARNING!!! If you enable this option, you MUST make sure that the 2141 module DOES NOT get stripped after being signed. This includes the 2142 debuginfo strip done by some packagers (such as rpmbuild) and 2143 inclusion into an initramfs that wants the module size reduced. 2144 2145config MODULE_SIG_FORCE 2146 bool "Require modules to be validly signed" 2147 depends on MODULE_SIG 2148 help 2149 Reject unsigned modules or signed modules for which we don't have a 2150 key. Without this, such modules will simply taint the kernel. 2151 2152config MODULE_SIG_ALL 2153 bool "Automatically sign all modules" 2154 default y 2155 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2156 help 2157 Sign all modules during make modules_install. Without this option, 2158 modules must be signed manually, using the scripts/sign-file tool. 2159 2160comment "Do not forget to sign required modules with scripts/sign-file" 2161 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL 2162 2163choice 2164 prompt "Which hash algorithm should modules be signed with?" 2165 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2166 help 2167 This determines which sort of hashing algorithm will be used during 2168 signature generation. This algorithm _must_ be built into the kernel 2169 directly so that signature verification can take place. It is not 2170 possible to load a signed module containing the algorithm to check 2171 the signature on that module. 2172 2173config MODULE_SIG_SHA1 2174 bool "Sign modules with SHA-1" 2175 select CRYPTO_SHA1 2176 2177config MODULE_SIG_SHA224 2178 bool "Sign modules with SHA-224" 2179 select CRYPTO_SHA256 2180 2181config MODULE_SIG_SHA256 2182 bool "Sign modules with SHA-256" 2183 select CRYPTO_SHA256 2184 2185config MODULE_SIG_SHA384 2186 bool "Sign modules with SHA-384" 2187 select CRYPTO_SHA512 2188 2189config MODULE_SIG_SHA512 2190 bool "Sign modules with SHA-512" 2191 select CRYPTO_SHA512 2192 2193endchoice 2194 2195config MODULE_SIG_HASH 2196 string 2197 depends on MODULE_SIG || IMA_APPRAISE_MODSIG 2198 default "sha1" if MODULE_SIG_SHA1 2199 default "sha224" if MODULE_SIG_SHA224 2200 default "sha256" if MODULE_SIG_SHA256 2201 default "sha384" if MODULE_SIG_SHA384 2202 default "sha512" if MODULE_SIG_SHA512 2203 2204choice 2205 prompt "Module compression mode" 2206 help 2207 This option allows you to choose the algorithm which will be used to 2208 compress modules when 'make modules_install' is run. (or, you can 2209 choose to not compress modules at all.) 2210 2211 External modules will also be compressed in the same way during the 2212 installation. 2213 2214 For modules inside an initrd or initramfs, it's more efficient to 2215 compress the whole initrd or initramfs instead. 2216 2217 This is fully compatible with signed modules. 2218 2219 Please note that the tool used to load modules needs to support the 2220 corresponding algorithm. module-init-tools MAY support gzip, and kmod 2221 MAY support gzip, xz and zstd. 2222 2223 Your build system needs to provide the appropriate compression tool 2224 to compress the modules. 2225 2226 If in doubt, select 'None'. 2227 2228config MODULE_COMPRESS_NONE 2229 bool "None" 2230 help 2231 Do not compress modules. The installed modules are suffixed 2232 with .ko. 2233 2234config MODULE_COMPRESS_GZIP 2235 bool "GZIP" 2236 help 2237 Compress modules with GZIP. The installed modules are suffixed 2238 with .ko.gz. 2239 2240config MODULE_COMPRESS_XZ 2241 bool "XZ" 2242 help 2243 Compress modules with XZ. The installed modules are suffixed 2244 with .ko.xz. 2245 2246config MODULE_COMPRESS_ZSTD 2247 bool "ZSTD" 2248 help 2249 Compress modules with ZSTD. The installed modules are suffixed 2250 with .ko.zst. 2251 2252endchoice 2253 2254config MODULE_ALLOW_MISSING_NAMESPACE_IMPORTS 2255 bool "Allow loading of modules with missing namespace imports" 2256 help 2257 Symbols exported with EXPORT_SYMBOL_NS*() are considered exported in 2258 a namespace. A module that makes use of a symbol exported with such a 2259 namespace is required to import the namespace via MODULE_IMPORT_NS(). 2260 There is no technical reason to enforce correct namespace imports, 2261 but it creates consistency between symbols defining namespaces and 2262 users importing namespaces they make use of. This option relaxes this 2263 requirement and lifts the enforcement when loading a module. 2264 2265 If unsure, say N. 2266 2267config MODPROBE_PATH 2268 string "Path to modprobe binary" 2269 default "/sbin/modprobe" 2270 help 2271 When kernel code requests a module, it does so by calling 2272 the "modprobe" userspace utility. This option allows you to 2273 set the path where that binary is found. This can be changed 2274 at runtime via the sysctl file 2275 /proc/sys/kernel/modprobe. Setting this to the empty string 2276 removes the kernel's ability to request modules (but 2277 userspace can still load modules explicitly). 2278 2279config TRIM_UNUSED_KSYMS 2280 bool "Trim unused exported kernel symbols" if EXPERT 2281 depends on !COMPILE_TEST 2282 help 2283 The kernel and some modules make many symbols available for 2284 other modules to use via EXPORT_SYMBOL() and variants. Depending 2285 on the set of modules being selected in your kernel configuration, 2286 many of those exported symbols might never be used. 2287 2288 This option allows for unused exported symbols to be dropped from 2289 the build. In turn, this provides the compiler more opportunities 2290 (especially when using LTO) for optimizing the code and reducing 2291 binary size. This might have some security advantages as well. 2292 2293 If unsure, or if you need to build out-of-tree modules, say N. 2294 2295config UNUSED_KSYMS_WHITELIST 2296 string "Whitelist of symbols to keep in ksymtab" 2297 depends on TRIM_UNUSED_KSYMS 2298 help 2299 By default, all unused exported symbols will be un-exported from the 2300 build when TRIM_UNUSED_KSYMS is selected. 2301 2302 UNUSED_KSYMS_WHITELIST allows to whitelist symbols that must be kept 2303 exported at all times, even in absence of in-tree users. The value to 2304 set here is the path to a text file containing the list of symbols, 2305 one per line. The path can be absolute, or relative to the kernel 2306 source tree. 2307 2308endif # MODULES 2309 2310config MODULES_TREE_LOOKUP 2311 def_bool y 2312 depends on PERF_EVENTS || TRACING || CFI_CLANG 2313 2314config INIT_ALL_POSSIBLE 2315 bool 2316 help 2317 Back when each arch used to define their own cpu_online_mask and 2318 cpu_possible_mask, some of them chose to initialize cpu_possible_mask 2319 with all 1s, and others with all 0s. When they were centralised, 2320 it was better to provide this option than to break all the archs 2321 and have several arch maintainers pursuing me down dark alleys. 2322 2323source "block/Kconfig" 2324 2325config PREEMPT_NOTIFIERS 2326 bool 2327 2328config PADATA 2329 depends on SMP 2330 bool 2331 2332config ASN1 2333 tristate 2334 help 2335 Build a simple ASN.1 grammar compiler that produces a bytecode output 2336 that can be interpreted by the ASN.1 stream decoder and used to 2337 inform it as to what tags are to be expected in a stream and what 2338 functions to call on what tags. 2339 2340source "kernel/Kconfig.locks" 2341 2342config ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE 2343 bool 2344 2345config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE 2346 bool 2347 2348# It may be useful for an architecture to override the definitions of the 2349# SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h> 2350# and the COMPAT_ variants in <linux/compat.h>, in particular to use a 2351# different calling convention for syscalls. They can also override the 2352# macros for not-implemented syscalls in kernel/sys_ni.c and 2353# kernel/time/posix-stubs.c. All these overrides need to be available in 2354# <asm/syscall_wrapper.h>. 2355config ARCH_HAS_SYSCALL_WRAPPER 2356 def_bool n 2357