1# SPDX-License-Identifier: GPL-2.0-only 2config DEFCONFIG_LIST 3 string 4 depends on !UML 5 option defconfig_list 6 default "/lib/modules/$(shell,uname -r)/.config" 7 default "/etc/kernel-config" 8 default "/boot/config-$(shell,uname -r)" 9 default ARCH_DEFCONFIG 10 default "arch/$(ARCH)/defconfig" 11 12config CC_IS_GCC 13 def_bool $(success,$(CC) --version | head -n 1 | grep -q gcc) 14 15config GCC_VERSION 16 int 17 default $(shell,$(srctree)/scripts/gcc-version.sh $(CC)) if CC_IS_GCC 18 default 0 19 20config CC_IS_CLANG 21 def_bool $(success,$(CC) --version | head -n 1 | grep -q clang) 22 23config CLANG_VERSION 24 int 25 default $(shell,$(srctree)/scripts/clang-version.sh $(CC)) 26 27config CC_CAN_LINK 28 def_bool $(success,$(srctree)/scripts/cc-can-link.sh $(CC)) 29 30config CC_HAS_ASM_GOTO 31 def_bool $(success,$(srctree)/scripts/gcc-goto.sh $(CC)) 32 33config TOOLS_SUPPORT_RELR 34 def_bool $(success,env "CC=$(CC)" "LD=$(LD)" "NM=$(NM)" "OBJCOPY=$(OBJCOPY)" $(srctree)/scripts/tools-support-relr.sh) 35 36config CC_HAS_ASM_INLINE 37 def_bool $(success,echo 'void foo(void) { asm inline (""); }' | $(CC) -x c - -c -o /dev/null) 38 39config CC_HAS_WARN_MAYBE_UNINITIALIZED 40 def_bool $(cc-option,-Wmaybe-uninitialized) 41 help 42 GCC >= 4.7 supports this option. 43 44config CC_DISABLE_WARN_MAYBE_UNINITIALIZED 45 bool 46 depends on CC_HAS_WARN_MAYBE_UNINITIALIZED 47 default CC_IS_GCC && GCC_VERSION < 40900 # unreliable for GCC < 4.9 48 help 49 GCC's -Wmaybe-uninitialized is not reliable by definition. 50 Lots of false positive warnings are produced in some cases. 51 52 If this option is enabled, -Wno-maybe-uninitialzed is passed 53 to the compiler to suppress maybe-uninitialized warnings. 54 55config CONSTRUCTORS 56 bool 57 58config IRQ_WORK 59 bool 60 61config BUILDTIME_EXTABLE_SORT 62 bool 63 64config THREAD_INFO_IN_TASK 65 bool 66 help 67 Select this to move thread_info off the stack into task_struct. To 68 make this work, an arch will need to remove all thread_info fields 69 except flags and fix any runtime bugs. 70 71 One subtle change that will be needed is to use try_get_task_stack() 72 and put_task_stack() in save_thread_stack_tsk() and get_wchan(). 73 74menu "General setup" 75 76config BROKEN 77 bool 78 79config BROKEN_ON_SMP 80 bool 81 depends on BROKEN || !SMP 82 default y 83 84config INIT_ENV_ARG_LIMIT 85 int 86 default 32 if !UML 87 default 128 if UML 88 help 89 Maximum of each of the number of arguments and environment 90 variables passed to init from the kernel command line. 91 92config COMPILE_TEST 93 bool "Compile also drivers which will not load" 94 depends on !UML 95 default n 96 help 97 Some drivers can be compiled on a different platform than they are 98 intended to be run on. Despite they cannot be loaded there (or even 99 when they load they cannot be used due to missing HW support), 100 developers still, opposing to distributors, might want to build such 101 drivers to compile-test them. 102 103 If you are a developer and want to build everything available, say Y 104 here. If you are a user/distributor, say N here to exclude useless 105 drivers to be distributed. 106 107config UAPI_HEADER_TEST 108 bool "Compile test UAPI headers" 109 depends on HEADERS_INSTALL && CC_CAN_LINK 110 help 111 Compile test headers exported to user-space to ensure they are 112 self-contained, i.e. compilable as standalone units. 113 114 If you are a developer or tester and want to ensure the exported 115 headers are self-contained, say Y here. Otherwise, choose N. 116 117config LOCALVERSION 118 string "Local version - append to kernel release" 119 help 120 Append an extra string to the end of your kernel version. 121 This will show up when you type uname, for example. 122 The string you set here will be appended after the contents of 123 any files with a filename matching localversion* in your 124 object and source tree, in that order. Your total string can 125 be a maximum of 64 characters. 126 127config LOCALVERSION_AUTO 128 bool "Automatically append version information to the version string" 129 default y 130 depends on !COMPILE_TEST 131 help 132 This will try to automatically determine if the current tree is a 133 release tree by looking for git tags that belong to the current 134 top of tree revision. 135 136 A string of the format -gxxxxxxxx will be added to the localversion 137 if a git-based tree is found. The string generated by this will be 138 appended after any matching localversion* files, and after the value 139 set in CONFIG_LOCALVERSION. 140 141 (The actual string used here is the first eight characters produced 142 by running the command: 143 144 $ git rev-parse --verify HEAD 145 146 which is done within the script "scripts/setlocalversion".) 147 148config BUILD_SALT 149 string "Build ID Salt" 150 default "" 151 help 152 The build ID is used to link binaries and their debug info. Setting 153 this option will use the value in the calculation of the build id. 154 This is mostly useful for distributions which want to ensure the 155 build is unique between builds. It's safe to leave the default. 156 157config HAVE_KERNEL_GZIP 158 bool 159 160config HAVE_KERNEL_BZIP2 161 bool 162 163config HAVE_KERNEL_LZMA 164 bool 165 166config HAVE_KERNEL_XZ 167 bool 168 169config HAVE_KERNEL_LZO 170 bool 171 172config HAVE_KERNEL_LZ4 173 bool 174 175config HAVE_KERNEL_UNCOMPRESSED 176 bool 177 178choice 179 prompt "Kernel compression mode" 180 default KERNEL_GZIP 181 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4 || HAVE_KERNEL_UNCOMPRESSED 182 help 183 The linux kernel is a kind of self-extracting executable. 184 Several compression algorithms are available, which differ 185 in efficiency, compression and decompression speed. 186 Compression speed is only relevant when building a kernel. 187 Decompression speed is relevant at each boot. 188 189 If you have any problems with bzip2 or lzma compressed 190 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 191 version of this functionality (bzip2 only), for 2.4, was 192 supplied by Christian Ludwig) 193 194 High compression options are mostly useful for users, who 195 are low on disk space (embedded systems), but for whom ram 196 size matters less. 197 198 If in doubt, select 'gzip' 199 200config KERNEL_GZIP 201 bool "Gzip" 202 depends on HAVE_KERNEL_GZIP 203 help 204 The old and tried gzip compression. It provides a good balance 205 between compression ratio and decompression speed. 206 207config KERNEL_BZIP2 208 bool "Bzip2" 209 depends on HAVE_KERNEL_BZIP2 210 help 211 Its compression ratio and speed is intermediate. 212 Decompression speed is slowest among the choices. The kernel 213 size is about 10% smaller with bzip2, in comparison to gzip. 214 Bzip2 uses a large amount of memory. For modern kernels you 215 will need at least 8MB RAM or more for booting. 216 217config KERNEL_LZMA 218 bool "LZMA" 219 depends on HAVE_KERNEL_LZMA 220 help 221 This compression algorithm's ratio is best. Decompression speed 222 is between gzip and bzip2. Compression is slowest. 223 The kernel size is about 33% smaller with LZMA in comparison to gzip. 224 225config KERNEL_XZ 226 bool "XZ" 227 depends on HAVE_KERNEL_XZ 228 help 229 XZ uses the LZMA2 algorithm and instruction set specific 230 BCJ filters which can improve compression ratio of executable 231 code. The size of the kernel is about 30% smaller with XZ in 232 comparison to gzip. On architectures for which there is a BCJ 233 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 234 will create a few percent smaller kernel than plain LZMA. 235 236 The speed is about the same as with LZMA: The decompression 237 speed of XZ is better than that of bzip2 but worse than gzip 238 and LZO. Compression is slow. 239 240config KERNEL_LZO 241 bool "LZO" 242 depends on HAVE_KERNEL_LZO 243 help 244 Its compression ratio is the poorest among the choices. The kernel 245 size is about 10% bigger than gzip; however its speed 246 (both compression and decompression) is the fastest. 247 248config KERNEL_LZ4 249 bool "LZ4" 250 depends on HAVE_KERNEL_LZ4 251 help 252 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding. 253 A preliminary version of LZ4 de/compression tool is available at 254 <https://code.google.com/p/lz4/>. 255 256 Its compression ratio is worse than LZO. The size of the kernel 257 is about 8% bigger than LZO. But the decompression speed is 258 faster than LZO. 259 260config KERNEL_UNCOMPRESSED 261 bool "None" 262 depends on HAVE_KERNEL_UNCOMPRESSED 263 help 264 Produce uncompressed kernel image. This option is usually not what 265 you want. It is useful for debugging the kernel in slow simulation 266 environments, where decompressing and moving the kernel is awfully 267 slow. This option allows early boot code to skip the decompressor 268 and jump right at uncompressed kernel image. 269 270endchoice 271 272config DEFAULT_HOSTNAME 273 string "Default hostname" 274 default "(none)" 275 help 276 This option determines the default system hostname before userspace 277 calls sethostname(2). The kernel traditionally uses "(none)" here, 278 but you may wish to use a different default here to make a minimal 279 system more usable with less configuration. 280 281# 282# For some reason microblaze and nios2 hard code SWAP=n. Hopefully we can 283# add proper SWAP support to them, in which case this can be remove. 284# 285config ARCH_NO_SWAP 286 bool 287 288config SWAP 289 bool "Support for paging of anonymous memory (swap)" 290 depends on MMU && BLOCK && !ARCH_NO_SWAP 291 default y 292 help 293 This option allows you to choose whether you want to have support 294 for so called swap devices or swap files in your kernel that are 295 used to provide more virtual memory than the actual RAM present 296 in your computer. If unsure say Y. 297 298config SYSVIPC 299 bool "System V IPC" 300 ---help--- 301 Inter Process Communication is a suite of library functions and 302 system calls which let processes (running programs) synchronize and 303 exchange information. It is generally considered to be a good thing, 304 and some programs won't run unless you say Y here. In particular, if 305 you want to run the DOS emulator dosemu under Linux (read the 306 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 307 you'll need to say Y here. 308 309 You can find documentation about IPC with "info ipc" and also in 310 section 6.4 of the Linux Programmer's Guide, available from 311 <http://www.tldp.org/guides.html>. 312 313config SYSVIPC_SYSCTL 314 bool 315 depends on SYSVIPC 316 depends on SYSCTL 317 default y 318 319config POSIX_MQUEUE 320 bool "POSIX Message Queues" 321 depends on NET 322 ---help--- 323 POSIX variant of message queues is a part of IPC. In POSIX message 324 queues every message has a priority which decides about succession 325 of receiving it by a process. If you want to compile and run 326 programs written e.g. for Solaris with use of its POSIX message 327 queues (functions mq_*) say Y here. 328 329 POSIX message queues are visible as a filesystem called 'mqueue' 330 and can be mounted somewhere if you want to do filesystem 331 operations on message queues. 332 333 If unsure, say Y. 334 335config POSIX_MQUEUE_SYSCTL 336 bool 337 depends on POSIX_MQUEUE 338 depends on SYSCTL 339 default y 340 341config CROSS_MEMORY_ATTACH 342 bool "Enable process_vm_readv/writev syscalls" 343 depends on MMU 344 default y 345 help 346 Enabling this option adds the system calls process_vm_readv and 347 process_vm_writev which allow a process with the correct privileges 348 to directly read from or write to another process' address space. 349 See the man page for more details. 350 351config USELIB 352 bool "uselib syscall" 353 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION 354 help 355 This option enables the uselib syscall, a system call used in the 356 dynamic linker from libc5 and earlier. glibc does not use this 357 system call. If you intend to run programs built on libc5 or 358 earlier, you may need to enable this syscall. Current systems 359 running glibc can safely disable this. 360 361config AUDIT 362 bool "Auditing support" 363 depends on NET 364 help 365 Enable auditing infrastructure that can be used with another 366 kernel subsystem, such as SELinux (which requires this for 367 logging of avc messages output). System call auditing is included 368 on architectures which support it. 369 370config HAVE_ARCH_AUDITSYSCALL 371 bool 372 373config AUDITSYSCALL 374 def_bool y 375 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL 376 select FSNOTIFY 377 378source "kernel/irq/Kconfig" 379source "kernel/time/Kconfig" 380source "kernel/Kconfig.preempt" 381 382menu "CPU/Task time and stats accounting" 383 384config VIRT_CPU_ACCOUNTING 385 bool 386 387choice 388 prompt "Cputime accounting" 389 default TICK_CPU_ACCOUNTING if !PPC64 390 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64 391 392# Kind of a stub config for the pure tick based cputime accounting 393config TICK_CPU_ACCOUNTING 394 bool "Simple tick based cputime accounting" 395 depends on !S390 && !NO_HZ_FULL 396 help 397 This is the basic tick based cputime accounting that maintains 398 statistics about user, system and idle time spent on per jiffies 399 granularity. 400 401 If unsure, say Y. 402 403config VIRT_CPU_ACCOUNTING_NATIVE 404 bool "Deterministic task and CPU time accounting" 405 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL 406 select VIRT_CPU_ACCOUNTING 407 help 408 Select this option to enable more accurate task and CPU time 409 accounting. This is done by reading a CPU counter on each 410 kernel entry and exit and on transitions within the kernel 411 between system, softirq and hardirq state, so there is a 412 small performance impact. In the case of s390 or IBM POWER > 5, 413 this also enables accounting of stolen time on logically-partitioned 414 systems. 415 416config VIRT_CPU_ACCOUNTING_GEN 417 bool "Full dynticks CPU time accounting" 418 depends on HAVE_CONTEXT_TRACKING 419 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN 420 depends on GENERIC_CLOCKEVENTS 421 select VIRT_CPU_ACCOUNTING 422 select CONTEXT_TRACKING 423 help 424 Select this option to enable task and CPU time accounting on full 425 dynticks systems. This accounting is implemented by watching every 426 kernel-user boundaries using the context tracking subsystem. 427 The accounting is thus performed at the expense of some significant 428 overhead. 429 430 For now this is only useful if you are working on the full 431 dynticks subsystem development. 432 433 If unsure, say N. 434 435endchoice 436 437config IRQ_TIME_ACCOUNTING 438 bool "Fine granularity task level IRQ time accounting" 439 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE 440 help 441 Select this option to enable fine granularity task irq time 442 accounting. This is done by reading a timestamp on each 443 transitions between softirq and hardirq state, so there can be a 444 small performance impact. 445 446 If in doubt, say N here. 447 448config HAVE_SCHED_AVG_IRQ 449 def_bool y 450 depends on IRQ_TIME_ACCOUNTING || PARAVIRT_TIME_ACCOUNTING 451 depends on SMP 452 453config BSD_PROCESS_ACCT 454 bool "BSD Process Accounting" 455 depends on MULTIUSER 456 help 457 If you say Y here, a user level program will be able to instruct the 458 kernel (via a special system call) to write process accounting 459 information to a file: whenever a process exits, information about 460 that process will be appended to the file by the kernel. The 461 information includes things such as creation time, owning user, 462 command name, memory usage, controlling terminal etc. (the complete 463 list is in the struct acct in <file:include/linux/acct.h>). It is 464 up to the user level program to do useful things with this 465 information. This is generally a good idea, so say Y. 466 467config BSD_PROCESS_ACCT_V3 468 bool "BSD Process Accounting version 3 file format" 469 depends on BSD_PROCESS_ACCT 470 default n 471 help 472 If you say Y here, the process accounting information is written 473 in a new file format that also logs the process IDs of each 474 process and its parent. Note that this file format is incompatible 475 with previous v0/v1/v2 file formats, so you will need updated tools 476 for processing it. A preliminary version of these tools is available 477 at <http://www.gnu.org/software/acct/>. 478 479config TASKSTATS 480 bool "Export task/process statistics through netlink" 481 depends on NET 482 depends on MULTIUSER 483 default n 484 help 485 Export selected statistics for tasks/processes through the 486 generic netlink interface. Unlike BSD process accounting, the 487 statistics are available during the lifetime of tasks/processes as 488 responses to commands. Like BSD accounting, they are sent to user 489 space on task exit. 490 491 Say N if unsure. 492 493config TASK_DELAY_ACCT 494 bool "Enable per-task delay accounting" 495 depends on TASKSTATS 496 select SCHED_INFO 497 help 498 Collect information on time spent by a task waiting for system 499 resources like cpu, synchronous block I/O completion and swapping 500 in pages. Such statistics can help in setting a task's priorities 501 relative to other tasks for cpu, io, rss limits etc. 502 503 Say N if unsure. 504 505config TASK_XACCT 506 bool "Enable extended accounting over taskstats" 507 depends on TASKSTATS 508 help 509 Collect extended task accounting data and send the data 510 to userland for processing over the taskstats interface. 511 512 Say N if unsure. 513 514config TASK_IO_ACCOUNTING 515 bool "Enable per-task storage I/O accounting" 516 depends on TASK_XACCT 517 help 518 Collect information on the number of bytes of storage I/O which this 519 task has caused. 520 521 Say N if unsure. 522 523config PSI 524 bool "Pressure stall information tracking" 525 help 526 Collect metrics that indicate how overcommitted the CPU, memory, 527 and IO capacity are in the system. 528 529 If you say Y here, the kernel will create /proc/pressure/ with the 530 pressure statistics files cpu, memory, and io. These will indicate 531 the share of walltime in which some or all tasks in the system are 532 delayed due to contention of the respective resource. 533 534 In kernels with cgroup support, cgroups (cgroup2 only) will 535 have cpu.pressure, memory.pressure, and io.pressure files, 536 which aggregate pressure stalls for the grouped tasks only. 537 538 For more details see Documentation/accounting/psi.rst. 539 540 Say N if unsure. 541 542config PSI_DEFAULT_DISABLED 543 bool "Require boot parameter to enable pressure stall information tracking" 544 default n 545 depends on PSI 546 help 547 If set, pressure stall information tracking will be disabled 548 per default but can be enabled through passing psi=1 on the 549 kernel commandline during boot. 550 551 This feature adds some code to the task wakeup and sleep 552 paths of the scheduler. The overhead is too low to affect 553 common scheduling-intense workloads in practice (such as 554 webservers, memcache), but it does show up in artificial 555 scheduler stress tests, such as hackbench. 556 557 If you are paranoid and not sure what the kernel will be 558 used for, say Y. 559 560 Say N if unsure. 561 562endmenu # "CPU/Task time and stats accounting" 563 564config CPU_ISOLATION 565 bool "CPU isolation" 566 depends on SMP || COMPILE_TEST 567 default y 568 help 569 Make sure that CPUs running critical tasks are not disturbed by 570 any source of "noise" such as unbound workqueues, timers, kthreads... 571 Unbound jobs get offloaded to housekeeping CPUs. This is driven by 572 the "isolcpus=" boot parameter. 573 574 Say Y if unsure. 575 576source "kernel/rcu/Kconfig" 577 578config BUILD_BIN2C 579 bool 580 default n 581 582config IKCONFIG 583 tristate "Kernel .config support" 584 ---help--- 585 This option enables the complete Linux kernel ".config" file 586 contents to be saved in the kernel. It provides documentation 587 of which kernel options are used in a running kernel or in an 588 on-disk kernel. This information can be extracted from the kernel 589 image file with the script scripts/extract-ikconfig and used as 590 input to rebuild the current kernel or to build another kernel. 591 It can also be extracted from a running kernel by reading 592 /proc/config.gz if enabled (below). 593 594config IKCONFIG_PROC 595 bool "Enable access to .config through /proc/config.gz" 596 depends on IKCONFIG && PROC_FS 597 ---help--- 598 This option enables access to the kernel configuration file 599 through /proc/config.gz. 600 601config IKHEADERS 602 tristate "Enable kernel headers through /sys/kernel/kheaders.tar.xz" 603 depends on SYSFS 604 help 605 This option enables access to the in-kernel headers that are generated during 606 the build process. These can be used to build eBPF tracing programs, 607 or similar programs. If you build the headers as a module, a module called 608 kheaders.ko is built which can be loaded on-demand to get access to headers. 609 610config LOG_BUF_SHIFT 611 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 612 range 12 25 613 default 17 614 depends on PRINTK 615 help 616 Select the minimal kernel log buffer size as a power of 2. 617 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config 618 parameter, see below. Any higher size also might be forced 619 by "log_buf_len" boot parameter. 620 621 Examples: 622 17 => 128 KB 623 16 => 64 KB 624 15 => 32 KB 625 14 => 16 KB 626 13 => 8 KB 627 12 => 4 KB 628 629config LOG_CPU_MAX_BUF_SHIFT 630 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)" 631 depends on SMP 632 range 0 21 633 default 12 if !BASE_SMALL 634 default 0 if BASE_SMALL 635 depends on PRINTK 636 help 637 This option allows to increase the default ring buffer size 638 according to the number of CPUs. The value defines the contribution 639 of each CPU as a power of 2. The used space is typically only few 640 lines however it might be much more when problems are reported, 641 e.g. backtraces. 642 643 The increased size means that a new buffer has to be allocated and 644 the original static one is unused. It makes sense only on systems 645 with more CPUs. Therefore this value is used only when the sum of 646 contributions is greater than the half of the default kernel ring 647 buffer as defined by LOG_BUF_SHIFT. The default values are set 648 so that more than 64 CPUs are needed to trigger the allocation. 649 650 Also this option is ignored when "log_buf_len" kernel parameter is 651 used as it forces an exact (power of two) size of the ring buffer. 652 653 The number of possible CPUs is used for this computation ignoring 654 hotplugging making the computation optimal for the worst case 655 scenario while allowing a simple algorithm to be used from bootup. 656 657 Examples shift values and their meaning: 658 17 => 128 KB for each CPU 659 16 => 64 KB for each CPU 660 15 => 32 KB for each CPU 661 14 => 16 KB for each CPU 662 13 => 8 KB for each CPU 663 12 => 4 KB for each CPU 664 665config PRINTK_SAFE_LOG_BUF_SHIFT 666 int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)" 667 range 10 21 668 default 13 669 depends on PRINTK 670 help 671 Select the size of an alternate printk per-CPU buffer where messages 672 printed from usafe contexts are temporary stored. One example would 673 be NMI messages, another one - printk recursion. The messages are 674 copied to the main log buffer in a safe context to avoid a deadlock. 675 The value defines the size as a power of 2. 676 677 Those messages are rare and limited. The largest one is when 678 a backtrace is printed. It usually fits into 4KB. Select 679 8KB if you want to be on the safe side. 680 681 Examples: 682 17 => 128 KB for each CPU 683 16 => 64 KB for each CPU 684 15 => 32 KB for each CPU 685 14 => 16 KB for each CPU 686 13 => 8 KB for each CPU 687 12 => 4 KB for each CPU 688 689# 690# Architectures with an unreliable sched_clock() should select this: 691# 692config HAVE_UNSTABLE_SCHED_CLOCK 693 bool 694 695config GENERIC_SCHED_CLOCK 696 bool 697 698menu "Scheduler features" 699 700config UCLAMP_TASK 701 bool "Enable utilization clamping for RT/FAIR tasks" 702 depends on CPU_FREQ_GOV_SCHEDUTIL 703 help 704 This feature enables the scheduler to track the clamped utilization 705 of each CPU based on RUNNABLE tasks scheduled on that CPU. 706 707 With this option, the user can specify the min and max CPU 708 utilization allowed for RUNNABLE tasks. The max utilization defines 709 the maximum frequency a task should use while the min utilization 710 defines the minimum frequency it should use. 711 712 Both min and max utilization clamp values are hints to the scheduler, 713 aiming at improving its frequency selection policy, but they do not 714 enforce or grant any specific bandwidth for tasks. 715 716 If in doubt, say N. 717 718config UCLAMP_BUCKETS_COUNT 719 int "Number of supported utilization clamp buckets" 720 range 5 20 721 default 5 722 depends on UCLAMP_TASK 723 help 724 Defines the number of clamp buckets to use. The range of each bucket 725 will be SCHED_CAPACITY_SCALE/UCLAMP_BUCKETS_COUNT. The higher the 726 number of clamp buckets the finer their granularity and the higher 727 the precision of clamping aggregation and tracking at run-time. 728 729 For example, with the minimum configuration value we will have 5 730 clamp buckets tracking 20% utilization each. A 25% boosted tasks will 731 be refcounted in the [20..39]% bucket and will set the bucket clamp 732 effective value to 25%. 733 If a second 30% boosted task should be co-scheduled on the same CPU, 734 that task will be refcounted in the same bucket of the first task and 735 it will boost the bucket clamp effective value to 30%. 736 The clamp effective value of a bucket is reset to its nominal value 737 (20% in the example above) when there are no more tasks refcounted in 738 that bucket. 739 740 An additional boost/capping margin can be added to some tasks. In the 741 example above the 25% task will be boosted to 30% until it exits the 742 CPU. If that should be considered not acceptable on certain systems, 743 it's always possible to reduce the margin by increasing the number of 744 clamp buckets to trade off used memory for run-time tracking 745 precision. 746 747 If in doubt, use the default value. 748 749endmenu 750 751# 752# For architectures that want to enable the support for NUMA-affine scheduler 753# balancing logic: 754# 755config ARCH_SUPPORTS_NUMA_BALANCING 756 bool 757 758# 759# For architectures that prefer to flush all TLBs after a number of pages 760# are unmapped instead of sending one IPI per page to flush. The architecture 761# must provide guarantees on what happens if a clean TLB cache entry is 762# written after the unmap. Details are in mm/rmap.c near the check for 763# should_defer_flush. The architecture should also consider if the full flush 764# and the refill costs are offset by the savings of sending fewer IPIs. 765config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH 766 bool 767 768# 769# For architectures that know their GCC __int128 support is sound 770# 771config ARCH_SUPPORTS_INT128 772 bool 773 774# For architectures that (ab)use NUMA to represent different memory regions 775# all cpu-local but of different latencies, such as SuperH. 776# 777config ARCH_WANT_NUMA_VARIABLE_LOCALITY 778 bool 779 780config NUMA_BALANCING 781 bool "Memory placement aware NUMA scheduler" 782 depends on ARCH_SUPPORTS_NUMA_BALANCING 783 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY 784 depends on SMP && NUMA && MIGRATION 785 help 786 This option adds support for automatic NUMA aware memory/task placement. 787 The mechanism is quite primitive and is based on migrating memory when 788 it has references to the node the task is running on. 789 790 This system will be inactive on UMA systems. 791 792config NUMA_BALANCING_DEFAULT_ENABLED 793 bool "Automatically enable NUMA aware memory/task placement" 794 default y 795 depends on NUMA_BALANCING 796 help 797 If set, automatic NUMA balancing will be enabled if running on a NUMA 798 machine. 799 800menuconfig CGROUPS 801 bool "Control Group support" 802 select KERNFS 803 help 804 This option adds support for grouping sets of processes together, for 805 use with process control subsystems such as Cpusets, CFS, memory 806 controls or device isolation. 807 See 808 - Documentation/scheduler/sched-design-CFS.rst (CFS) 809 - Documentation/admin-guide/cgroup-v1/ (features for grouping, isolation 810 and resource control) 811 812 Say N if unsure. 813 814if CGROUPS 815 816config PAGE_COUNTER 817 bool 818 819config MEMCG 820 bool "Memory controller" 821 select PAGE_COUNTER 822 select EVENTFD 823 help 824 Provides control over the memory footprint of tasks in a cgroup. 825 826config MEMCG_SWAP 827 bool "Swap controller" 828 depends on MEMCG && SWAP 829 help 830 Provides control over the swap space consumed by tasks in a cgroup. 831 832config MEMCG_SWAP_ENABLED 833 bool "Swap controller enabled by default" 834 depends on MEMCG_SWAP 835 default y 836 help 837 Memory Resource Controller Swap Extension comes with its price in 838 a bigger memory consumption. General purpose distribution kernels 839 which want to enable the feature but keep it disabled by default 840 and let the user enable it by swapaccount=1 boot command line 841 parameter should have this option unselected. 842 For those who want to have the feature enabled by default should 843 select this option (if, for some reason, they need to disable it 844 then swapaccount=0 does the trick). 845 846config MEMCG_KMEM 847 bool 848 depends on MEMCG && !SLOB 849 default y 850 851config BLK_CGROUP 852 bool "IO controller" 853 depends on BLOCK 854 default n 855 ---help--- 856 Generic block IO controller cgroup interface. This is the common 857 cgroup interface which should be used by various IO controlling 858 policies. 859 860 Currently, CFQ IO scheduler uses it to recognize task groups and 861 control disk bandwidth allocation (proportional time slice allocation) 862 to such task groups. It is also used by bio throttling logic in 863 block layer to implement upper limit in IO rates on a device. 864 865 This option only enables generic Block IO controller infrastructure. 866 One needs to also enable actual IO controlling logic/policy. For 867 enabling proportional weight division of disk bandwidth in CFQ, set 868 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 869 CONFIG_BLK_DEV_THROTTLING=y. 870 871 See Documentation/admin-guide/cgroup-v1/blkio-controller.rst for more information. 872 873config CGROUP_WRITEBACK 874 bool 875 depends on MEMCG && BLK_CGROUP 876 default y 877 878menuconfig CGROUP_SCHED 879 bool "CPU controller" 880 default n 881 help 882 This feature lets CPU scheduler recognize task groups and control CPU 883 bandwidth allocation to such task groups. It uses cgroups to group 884 tasks. 885 886if CGROUP_SCHED 887config FAIR_GROUP_SCHED 888 bool "Group scheduling for SCHED_OTHER" 889 depends on CGROUP_SCHED 890 default CGROUP_SCHED 891 892config CFS_BANDWIDTH 893 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 894 depends on FAIR_GROUP_SCHED 895 default n 896 help 897 This option allows users to define CPU bandwidth rates (limits) for 898 tasks running within the fair group scheduler. Groups with no limit 899 set are considered to be unconstrained and will run with no 900 restriction. 901 See Documentation/scheduler/sched-bwc.rst for more information. 902 903config RT_GROUP_SCHED 904 bool "Group scheduling for SCHED_RR/FIFO" 905 depends on CGROUP_SCHED 906 default n 907 help 908 This feature lets you explicitly allocate real CPU bandwidth 909 to task groups. If enabled, it will also make it impossible to 910 schedule realtime tasks for non-root users until you allocate 911 realtime bandwidth for them. 912 See Documentation/scheduler/sched-rt-group.rst for more information. 913 914endif #CGROUP_SCHED 915 916config UCLAMP_TASK_GROUP 917 bool "Utilization clamping per group of tasks" 918 depends on CGROUP_SCHED 919 depends on UCLAMP_TASK 920 default n 921 help 922 This feature enables the scheduler to track the clamped utilization 923 of each CPU based on RUNNABLE tasks currently scheduled on that CPU. 924 925 When this option is enabled, the user can specify a min and max 926 CPU bandwidth which is allowed for each single task in a group. 927 The max bandwidth allows to clamp the maximum frequency a task 928 can use, while the min bandwidth allows to define a minimum 929 frequency a task will always use. 930 931 When task group based utilization clamping is enabled, an eventually 932 specified task-specific clamp value is constrained by the cgroup 933 specified clamp value. Both minimum and maximum task clamping cannot 934 be bigger than the corresponding clamping defined at task group level. 935 936 If in doubt, say N. 937 938config CGROUP_PIDS 939 bool "PIDs controller" 940 help 941 Provides enforcement of process number limits in the scope of a 942 cgroup. Any attempt to fork more processes than is allowed in the 943 cgroup will fail. PIDs are fundamentally a global resource because it 944 is fairly trivial to reach PID exhaustion before you reach even a 945 conservative kmemcg limit. As a result, it is possible to grind a 946 system to halt without being limited by other cgroup policies. The 947 PIDs controller is designed to stop this from happening. 948 949 It should be noted that organisational operations (such as attaching 950 to a cgroup hierarchy) will *not* be blocked by the PIDs controller, 951 since the PIDs limit only affects a process's ability to fork, not to 952 attach to a cgroup. 953 954config CGROUP_RDMA 955 bool "RDMA controller" 956 help 957 Provides enforcement of RDMA resources defined by IB stack. 958 It is fairly easy for consumers to exhaust RDMA resources, which 959 can result into resource unavailability to other consumers. 960 RDMA controller is designed to stop this from happening. 961 Attaching processes with active RDMA resources to the cgroup 962 hierarchy is allowed even if can cross the hierarchy's limit. 963 964config CGROUP_FREEZER 965 bool "Freezer controller" 966 help 967 Provides a way to freeze and unfreeze all tasks in a 968 cgroup. 969 970 This option affects the ORIGINAL cgroup interface. The cgroup2 memory 971 controller includes important in-kernel memory consumers per default. 972 973 If you're using cgroup2, say N. 974 975config CGROUP_HUGETLB 976 bool "HugeTLB controller" 977 depends on HUGETLB_PAGE 978 select PAGE_COUNTER 979 default n 980 help 981 Provides a cgroup controller for HugeTLB pages. 982 When you enable this, you can put a per cgroup limit on HugeTLB usage. 983 The limit is enforced during page fault. Since HugeTLB doesn't 984 support page reclaim, enforcing the limit at page fault time implies 985 that, the application will get SIGBUS signal if it tries to access 986 HugeTLB pages beyond its limit. This requires the application to know 987 beforehand how much HugeTLB pages it would require for its use. The 988 control group is tracked in the third page lru pointer. This means 989 that we cannot use the controller with huge page less than 3 pages. 990 991config CPUSETS 992 bool "Cpuset controller" 993 depends on SMP 994 help 995 This option will let you create and manage CPUSETs which 996 allow dynamically partitioning a system into sets of CPUs and 997 Memory Nodes and assigning tasks to run only within those sets. 998 This is primarily useful on large SMP or NUMA systems. 999 1000 Say N if unsure. 1001 1002config PROC_PID_CPUSET 1003 bool "Include legacy /proc/<pid>/cpuset file" 1004 depends on CPUSETS 1005 default y 1006 1007config CGROUP_DEVICE 1008 bool "Device controller" 1009 help 1010 Provides a cgroup controller implementing whitelists for 1011 devices which a process in the cgroup can mknod or open. 1012 1013config CGROUP_CPUACCT 1014 bool "Simple CPU accounting controller" 1015 help 1016 Provides a simple controller for monitoring the 1017 total CPU consumed by the tasks in a cgroup. 1018 1019config CGROUP_PERF 1020 bool "Perf controller" 1021 depends on PERF_EVENTS 1022 help 1023 This option extends the perf per-cpu mode to restrict monitoring 1024 to threads which belong to the cgroup specified and run on the 1025 designated cpu. 1026 1027 Say N if unsure. 1028 1029config CGROUP_BPF 1030 bool "Support for eBPF programs attached to cgroups" 1031 depends on BPF_SYSCALL 1032 select SOCK_CGROUP_DATA 1033 help 1034 Allow attaching eBPF programs to a cgroup using the bpf(2) 1035 syscall command BPF_PROG_ATTACH. 1036 1037 In which context these programs are accessed depends on the type 1038 of attachment. For instance, programs that are attached using 1039 BPF_CGROUP_INET_INGRESS will be executed on the ingress path of 1040 inet sockets. 1041 1042config CGROUP_DEBUG 1043 bool "Debug controller" 1044 default n 1045 depends on DEBUG_KERNEL 1046 help 1047 This option enables a simple controller that exports 1048 debugging information about the cgroups framework. This 1049 controller is for control cgroup debugging only. Its 1050 interfaces are not stable. 1051 1052 Say N. 1053 1054config SOCK_CGROUP_DATA 1055 bool 1056 default n 1057 1058endif # CGROUPS 1059 1060menuconfig NAMESPACES 1061 bool "Namespaces support" if EXPERT 1062 depends on MULTIUSER 1063 default !EXPERT 1064 help 1065 Provides the way to make tasks work with different objects using 1066 the same id. For example same IPC id may refer to different objects 1067 or same user id or pid may refer to different tasks when used in 1068 different namespaces. 1069 1070if NAMESPACES 1071 1072config UTS_NS 1073 bool "UTS namespace" 1074 default y 1075 help 1076 In this namespace tasks see different info provided with the 1077 uname() system call 1078 1079config IPC_NS 1080 bool "IPC namespace" 1081 depends on (SYSVIPC || POSIX_MQUEUE) 1082 default y 1083 help 1084 In this namespace tasks work with IPC ids which correspond to 1085 different IPC objects in different namespaces. 1086 1087config USER_NS 1088 bool "User namespace" 1089 default n 1090 help 1091 This allows containers, i.e. vservers, to use user namespaces 1092 to provide different user info for different servers. 1093 1094 When user namespaces are enabled in the kernel it is 1095 recommended that the MEMCG option also be enabled and that 1096 user-space use the memory control groups to limit the amount 1097 of memory a memory unprivileged users can use. 1098 1099 If unsure, say N. 1100 1101config PID_NS 1102 bool "PID Namespaces" 1103 default y 1104 help 1105 Support process id namespaces. This allows having multiple 1106 processes with the same pid as long as they are in different 1107 pid namespaces. This is a building block of containers. 1108 1109config NET_NS 1110 bool "Network namespace" 1111 depends on NET 1112 default y 1113 help 1114 Allow user space to create what appear to be multiple instances 1115 of the network stack. 1116 1117endif # NAMESPACES 1118 1119config CHECKPOINT_RESTORE 1120 bool "Checkpoint/restore support" 1121 select PROC_CHILDREN 1122 default n 1123 help 1124 Enables additional kernel features in a sake of checkpoint/restore. 1125 In particular it adds auxiliary prctl codes to setup process text, 1126 data and heap segment sizes, and a few additional /proc filesystem 1127 entries. 1128 1129 If unsure, say N here. 1130 1131config SCHED_AUTOGROUP 1132 bool "Automatic process group scheduling" 1133 select CGROUPS 1134 select CGROUP_SCHED 1135 select FAIR_GROUP_SCHED 1136 help 1137 This option optimizes the scheduler for common desktop workloads by 1138 automatically creating and populating task groups. This separation 1139 of workloads isolates aggressive CPU burners (like build jobs) from 1140 desktop applications. Task group autogeneration is currently based 1141 upon task session. 1142 1143config SYSFS_DEPRECATED 1144 bool "Enable deprecated sysfs features to support old userspace tools" 1145 depends on SYSFS 1146 default n 1147 help 1148 This option adds code that switches the layout of the "block" class 1149 devices, to not show up in /sys/class/block/, but only in 1150 /sys/block/. 1151 1152 This switch is only active when the sysfs.deprecated=1 boot option is 1153 passed or the SYSFS_DEPRECATED_V2 option is set. 1154 1155 This option allows new kernels to run on old distributions and tools, 1156 which might get confused by /sys/class/block/. Since 2007/2008 all 1157 major distributions and tools handle this just fine. 1158 1159 Recent distributions and userspace tools after 2009/2010 depend on 1160 the existence of /sys/class/block/, and will not work with this 1161 option enabled. 1162 1163 Only if you are using a new kernel on an old distribution, you might 1164 need to say Y here. 1165 1166config SYSFS_DEPRECATED_V2 1167 bool "Enable deprecated sysfs features by default" 1168 default n 1169 depends on SYSFS 1170 depends on SYSFS_DEPRECATED 1171 help 1172 Enable deprecated sysfs by default. 1173 1174 See the CONFIG_SYSFS_DEPRECATED option for more details about this 1175 option. 1176 1177 Only if you are using a new kernel on an old distribution, you might 1178 need to say Y here. Even then, odds are you would not need it 1179 enabled, you can always pass the boot option if absolutely necessary. 1180 1181config RELAY 1182 bool "Kernel->user space relay support (formerly relayfs)" 1183 select IRQ_WORK 1184 help 1185 This option enables support for relay interface support in 1186 certain file systems (such as debugfs). 1187 It is designed to provide an efficient mechanism for tools and 1188 facilities to relay large amounts of data from kernel space to 1189 user space. 1190 1191 If unsure, say N. 1192 1193config BLK_DEV_INITRD 1194 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1195 help 1196 The initial RAM filesystem is a ramfs which is loaded by the 1197 boot loader (loadlin or lilo) and that is mounted as root 1198 before the normal boot procedure. It is typically used to 1199 load modules needed to mount the "real" root file system, 1200 etc. See <file:Documentation/admin-guide/initrd.rst> for details. 1201 1202 If RAM disk support (BLK_DEV_RAM) is also included, this 1203 also enables initial RAM disk (initrd) support and adds 1204 15 Kbytes (more on some other architectures) to the kernel size. 1205 1206 If unsure say Y. 1207 1208if BLK_DEV_INITRD 1209 1210source "usr/Kconfig" 1211 1212endif 1213 1214choice 1215 prompt "Compiler optimization level" 1216 default CC_OPTIMIZE_FOR_PERFORMANCE 1217 1218config CC_OPTIMIZE_FOR_PERFORMANCE 1219 bool "Optimize for performance (-O2)" 1220 help 1221 This is the default optimization level for the kernel, building 1222 with the "-O2" compiler flag for best performance and most 1223 helpful compile-time warnings. 1224 1225config CC_OPTIMIZE_FOR_PERFORMANCE_O3 1226 bool "Optimize more for performance (-O3)" 1227 depends on ARC 1228 imply CC_DISABLE_WARN_MAYBE_UNINITIALIZED # avoid false positives 1229 help 1230 Choosing this option will pass "-O3" to your compiler to optimize 1231 the kernel yet more for performance. 1232 1233config CC_OPTIMIZE_FOR_SIZE 1234 bool "Optimize for size (-Os)" 1235 imply CC_DISABLE_WARN_MAYBE_UNINITIALIZED # avoid false positives 1236 help 1237 Choosing this option will pass "-Os" to your compiler resulting 1238 in a smaller kernel. 1239 1240endchoice 1241 1242config HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1243 bool 1244 help 1245 This requires that the arch annotates or otherwise protects 1246 its external entry points from being discarded. Linker scripts 1247 must also merge .text.*, .data.*, and .bss.* correctly into 1248 output sections. Care must be taken not to pull in unrelated 1249 sections (e.g., '.text.init'). Typically '.' in section names 1250 is used to distinguish them from label names / C identifiers. 1251 1252config LD_DEAD_CODE_DATA_ELIMINATION 1253 bool "Dead code and data elimination (EXPERIMENTAL)" 1254 depends on HAVE_LD_DEAD_CODE_DATA_ELIMINATION 1255 depends on EXPERT 1256 depends on !(FUNCTION_TRACER && CC_IS_GCC && GCC_VERSION < 40800) 1257 depends on $(cc-option,-ffunction-sections -fdata-sections) 1258 depends on $(ld-option,--gc-sections) 1259 help 1260 Enable this if you want to do dead code and data elimination with 1261 the linker by compiling with -ffunction-sections -fdata-sections, 1262 and linking with --gc-sections. 1263 1264 This can reduce on disk and in-memory size of the kernel 1265 code and static data, particularly for small configs and 1266 on small systems. This has the possibility of introducing 1267 silently broken kernel if the required annotations are not 1268 present. This option is not well tested yet, so use at your 1269 own risk. 1270 1271config SYSCTL 1272 bool 1273 1274config HAVE_UID16 1275 bool 1276 1277config SYSCTL_EXCEPTION_TRACE 1278 bool 1279 help 1280 Enable support for /proc/sys/debug/exception-trace. 1281 1282config SYSCTL_ARCH_UNALIGN_NO_WARN 1283 bool 1284 help 1285 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap 1286 Allows arch to define/use @no_unaligned_warning to possibly warn 1287 about unaligned access emulation going on under the hood. 1288 1289config SYSCTL_ARCH_UNALIGN_ALLOW 1290 bool 1291 help 1292 Enable support for /proc/sys/kernel/unaligned-trap 1293 Allows arches to define/use @unaligned_enabled to runtime toggle 1294 the unaligned access emulation. 1295 see arch/parisc/kernel/unaligned.c for reference 1296 1297config HAVE_PCSPKR_PLATFORM 1298 bool 1299 1300# interpreter that classic socket filters depend on 1301config BPF 1302 bool 1303 1304menuconfig EXPERT 1305 bool "Configure standard kernel features (expert users)" 1306 # Unhide debug options, to make the on-by-default options visible 1307 select DEBUG_KERNEL 1308 help 1309 This option allows certain base kernel options and settings 1310 to be disabled or tweaked. This is for specialized 1311 environments which can tolerate a "non-standard" kernel. 1312 Only use this if you really know what you are doing. 1313 1314config UID16 1315 bool "Enable 16-bit UID system calls" if EXPERT 1316 depends on HAVE_UID16 && MULTIUSER 1317 default y 1318 help 1319 This enables the legacy 16-bit UID syscall wrappers. 1320 1321config MULTIUSER 1322 bool "Multiple users, groups and capabilities support" if EXPERT 1323 default y 1324 help 1325 This option enables support for non-root users, groups and 1326 capabilities. 1327 1328 If you say N here, all processes will run with UID 0, GID 0, and all 1329 possible capabilities. Saying N here also compiles out support for 1330 system calls related to UIDs, GIDs, and capabilities, such as setuid, 1331 setgid, and capset. 1332 1333 If unsure, say Y here. 1334 1335config SGETMASK_SYSCALL 1336 bool "sgetmask/ssetmask syscalls support" if EXPERT 1337 def_bool PARISC || M68K || PPC || MIPS || X86 || SPARC || MICROBLAZE || SUPERH 1338 ---help--- 1339 sys_sgetmask and sys_ssetmask are obsolete system calls 1340 no longer supported in libc but still enabled by default in some 1341 architectures. 1342 1343 If unsure, leave the default option here. 1344 1345config SYSFS_SYSCALL 1346 bool "Sysfs syscall support" if EXPERT 1347 default y 1348 ---help--- 1349 sys_sysfs is an obsolete system call no longer supported in libc. 1350 Note that disabling this option is more secure but might break 1351 compatibility with some systems. 1352 1353 If unsure say Y here. 1354 1355config SYSCTL_SYSCALL 1356 bool "Sysctl syscall support" if EXPERT 1357 depends on PROC_SYSCTL 1358 default n 1359 select SYSCTL 1360 ---help--- 1361 sys_sysctl uses binary paths that have been found challenging 1362 to properly maintain and use. The interface in /proc/sys 1363 using paths with ascii names is now the primary path to this 1364 information. 1365 1366 Almost nothing using the binary sysctl interface so if you are 1367 trying to save some space it is probably safe to disable this, 1368 making your kernel marginally smaller. 1369 1370 If unsure say N here. 1371 1372config FHANDLE 1373 bool "open by fhandle syscalls" if EXPERT 1374 select EXPORTFS 1375 default y 1376 help 1377 If you say Y here, a user level program will be able to map 1378 file names to handle and then later use the handle for 1379 different file system operations. This is useful in implementing 1380 userspace file servers, which now track files using handles instead 1381 of names. The handle would remain the same even if file names 1382 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 1383 syscalls. 1384 1385config POSIX_TIMERS 1386 bool "Posix Clocks & timers" if EXPERT 1387 default y 1388 help 1389 This includes native support for POSIX timers to the kernel. 1390 Some embedded systems have no use for them and therefore they 1391 can be configured out to reduce the size of the kernel image. 1392 1393 When this option is disabled, the following syscalls won't be 1394 available: timer_create, timer_gettime: timer_getoverrun, 1395 timer_settime, timer_delete, clock_adjtime, getitimer, 1396 setitimer, alarm. Furthermore, the clock_settime, clock_gettime, 1397 clock_getres and clock_nanosleep syscalls will be limited to 1398 CLOCK_REALTIME, CLOCK_MONOTONIC and CLOCK_BOOTTIME only. 1399 1400 If unsure say y. 1401 1402config PRINTK 1403 default y 1404 bool "Enable support for printk" if EXPERT 1405 select IRQ_WORK 1406 help 1407 This option enables normal printk support. Removing it 1408 eliminates most of the message strings from the kernel image 1409 and makes the kernel more or less silent. As this makes it 1410 very difficult to diagnose system problems, saying N here is 1411 strongly discouraged. 1412 1413config PRINTK_NMI 1414 def_bool y 1415 depends on PRINTK 1416 depends on HAVE_NMI 1417 1418config BUG 1419 bool "BUG() support" if EXPERT 1420 default y 1421 help 1422 Disabling this option eliminates support for BUG and WARN, reducing 1423 the size of your kernel image and potentially quietly ignoring 1424 numerous fatal conditions. You should only consider disabling this 1425 option for embedded systems with no facilities for reporting errors. 1426 Just say Y. 1427 1428config ELF_CORE 1429 depends on COREDUMP 1430 default y 1431 bool "Enable ELF core dumps" if EXPERT 1432 help 1433 Enable support for generating core dumps. Disabling saves about 4k. 1434 1435 1436config PCSPKR_PLATFORM 1437 bool "Enable PC-Speaker support" if EXPERT 1438 depends on HAVE_PCSPKR_PLATFORM 1439 select I8253_LOCK 1440 default y 1441 help 1442 This option allows to disable the internal PC-Speaker 1443 support, saving some memory. 1444 1445config BASE_FULL 1446 default y 1447 bool "Enable full-sized data structures for core" if EXPERT 1448 help 1449 Disabling this option reduces the size of miscellaneous core 1450 kernel data structures. This saves memory on small machines, 1451 but may reduce performance. 1452 1453config FUTEX 1454 bool "Enable futex support" if EXPERT 1455 default y 1456 imply RT_MUTEXES 1457 help 1458 Disabling this option will cause the kernel to be built without 1459 support for "fast userspace mutexes". The resulting kernel may not 1460 run glibc-based applications correctly. 1461 1462config FUTEX_PI 1463 bool 1464 depends on FUTEX && RT_MUTEXES 1465 default y 1466 1467config HAVE_FUTEX_CMPXCHG 1468 bool 1469 depends on FUTEX 1470 help 1471 Architectures should select this if futex_atomic_cmpxchg_inatomic() 1472 is implemented and always working. This removes a couple of runtime 1473 checks. 1474 1475config EPOLL 1476 bool "Enable eventpoll support" if EXPERT 1477 default y 1478 help 1479 Disabling this option will cause the kernel to be built without 1480 support for epoll family of system calls. 1481 1482config SIGNALFD 1483 bool "Enable signalfd() system call" if EXPERT 1484 default y 1485 help 1486 Enable the signalfd() system call that allows to receive signals 1487 on a file descriptor. 1488 1489 If unsure, say Y. 1490 1491config TIMERFD 1492 bool "Enable timerfd() system call" if EXPERT 1493 default y 1494 help 1495 Enable the timerfd() system call that allows to receive timer 1496 events on a file descriptor. 1497 1498 If unsure, say Y. 1499 1500config EVENTFD 1501 bool "Enable eventfd() system call" if EXPERT 1502 default y 1503 help 1504 Enable the eventfd() system call that allows to receive both 1505 kernel notification (ie. KAIO) or userspace notifications. 1506 1507 If unsure, say Y. 1508 1509config SHMEM 1510 bool "Use full shmem filesystem" if EXPERT 1511 default y 1512 depends on MMU 1513 help 1514 The shmem is an internal filesystem used to manage shared memory. 1515 It is backed by swap and manages resource limits. It is also exported 1516 to userspace as tmpfs if TMPFS is enabled. Disabling this 1517 option replaces shmem and tmpfs with the much simpler ramfs code, 1518 which may be appropriate on small systems without swap. 1519 1520config AIO 1521 bool "Enable AIO support" if EXPERT 1522 default y 1523 help 1524 This option enables POSIX asynchronous I/O which may by used 1525 by some high performance threaded applications. Disabling 1526 this option saves about 7k. 1527 1528config IO_URING 1529 bool "Enable IO uring support" if EXPERT 1530 select ANON_INODES 1531 default y 1532 help 1533 This option enables support for the io_uring interface, enabling 1534 applications to submit and complete IO through submission and 1535 completion rings that are shared between the kernel and application. 1536 1537config ADVISE_SYSCALLS 1538 bool "Enable madvise/fadvise syscalls" if EXPERT 1539 default y 1540 help 1541 This option enables the madvise and fadvise syscalls, used by 1542 applications to advise the kernel about their future memory or file 1543 usage, improving performance. If building an embedded system where no 1544 applications use these syscalls, you can disable this option to save 1545 space. 1546 1547config MEMBARRIER 1548 bool "Enable membarrier() system call" if EXPERT 1549 default y 1550 help 1551 Enable the membarrier() system call that allows issuing memory 1552 barriers across all running threads, which can be used to distribute 1553 the cost of user-space memory barriers asymmetrically by transforming 1554 pairs of memory barriers into pairs consisting of membarrier() and a 1555 compiler barrier. 1556 1557 If unsure, say Y. 1558 1559config KALLSYMS 1560 bool "Load all symbols for debugging/ksymoops" if EXPERT 1561 default y 1562 help 1563 Say Y here to let the kernel print out symbolic crash information and 1564 symbolic stack backtraces. This increases the size of the kernel 1565 somewhat, as all symbols have to be loaded into the kernel image. 1566 1567config KALLSYMS_ALL 1568 bool "Include all symbols in kallsyms" 1569 depends on DEBUG_KERNEL && KALLSYMS 1570 help 1571 Normally kallsyms only contains the symbols of functions for nicer 1572 OOPS messages and backtraces (i.e., symbols from the text and inittext 1573 sections). This is sufficient for most cases. And only in very rare 1574 cases (e.g., when a debugger is used) all symbols are required (e.g., 1575 names of variables from the data sections, etc). 1576 1577 This option makes sure that all symbols are loaded into the kernel 1578 image (i.e., symbols from all sections) in cost of increased kernel 1579 size (depending on the kernel configuration, it may be 300KiB or 1580 something like this). 1581 1582 Say N unless you really need all symbols. 1583 1584config KALLSYMS_ABSOLUTE_PERCPU 1585 bool 1586 depends on KALLSYMS 1587 default X86_64 && SMP 1588 1589config KALLSYMS_BASE_RELATIVE 1590 bool 1591 depends on KALLSYMS 1592 default !IA64 1593 help 1594 Instead of emitting them as absolute values in the native word size, 1595 emit the symbol references in the kallsyms table as 32-bit entries, 1596 each containing a relative value in the range [base, base + U32_MAX] 1597 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either 1598 an absolute value in the range [0, S32_MAX] or a relative value in the 1599 range [base, base + S32_MAX], where base is the lowest relative symbol 1600 address encountered in the image. 1601 1602 On 64-bit builds, this reduces the size of the address table by 50%, 1603 but more importantly, it results in entries whose values are build 1604 time constants, and no relocation pass is required at runtime to fix 1605 up the entries based on the runtime load address of the kernel. 1606 1607# end of the "standard kernel features (expert users)" menu 1608 1609# syscall, maps, verifier 1610config BPF_SYSCALL 1611 bool "Enable bpf() system call" 1612 select BPF 1613 select IRQ_WORK 1614 default n 1615 help 1616 Enable the bpf() system call that allows to manipulate eBPF 1617 programs and maps via file descriptors. 1618 1619config BPF_JIT_ALWAYS_ON 1620 bool "Permanently enable BPF JIT and remove BPF interpreter" 1621 depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT 1622 help 1623 Enables BPF JIT and removes BPF interpreter to avoid 1624 speculative execution of BPF instructions by the interpreter 1625 1626config USERFAULTFD 1627 bool "Enable userfaultfd() system call" 1628 depends on MMU 1629 help 1630 Enable the userfaultfd() system call that allows to intercept and 1631 handle page faults in userland. 1632 1633config ARCH_HAS_MEMBARRIER_CALLBACKS 1634 bool 1635 1636config ARCH_HAS_MEMBARRIER_SYNC_CORE 1637 bool 1638 1639config RSEQ 1640 bool "Enable rseq() system call" if EXPERT 1641 default y 1642 depends on HAVE_RSEQ 1643 select MEMBARRIER 1644 help 1645 Enable the restartable sequences system call. It provides a 1646 user-space cache for the current CPU number value, which 1647 speeds up getting the current CPU number from user-space, 1648 as well as an ABI to speed up user-space operations on 1649 per-CPU data. 1650 1651 If unsure, say Y. 1652 1653config DEBUG_RSEQ 1654 default n 1655 bool "Enabled debugging of rseq() system call" if EXPERT 1656 depends on RSEQ && DEBUG_KERNEL 1657 help 1658 Enable extra debugging checks for the rseq system call. 1659 1660 If unsure, say N. 1661 1662config EMBEDDED 1663 bool "Embedded system" 1664 option allnoconfig_y 1665 select EXPERT 1666 help 1667 This option should be enabled if compiling the kernel for 1668 an embedded system so certain expert options are available 1669 for configuration. 1670 1671config HAVE_PERF_EVENTS 1672 bool 1673 help 1674 See tools/perf/design.txt for details. 1675 1676config PERF_USE_VMALLOC 1677 bool 1678 help 1679 See tools/perf/design.txt for details 1680 1681config PC104 1682 bool "PC/104 support" if EXPERT 1683 help 1684 Expose PC/104 form factor device drivers and options available for 1685 selection and configuration. Enable this option if your target 1686 machine has a PC/104 bus. 1687 1688menu "Kernel Performance Events And Counters" 1689 1690config PERF_EVENTS 1691 bool "Kernel performance events and counters" 1692 default y if PROFILING 1693 depends on HAVE_PERF_EVENTS 1694 select IRQ_WORK 1695 select SRCU 1696 help 1697 Enable kernel support for various performance events provided 1698 by software and hardware. 1699 1700 Software events are supported either built-in or via the 1701 use of generic tracepoints. 1702 1703 Most modern CPUs support performance events via performance 1704 counter registers. These registers count the number of certain 1705 types of hw events: such as instructions executed, cachemisses 1706 suffered, or branches mis-predicted - without slowing down the 1707 kernel or applications. These registers can also trigger interrupts 1708 when a threshold number of events have passed - and can thus be 1709 used to profile the code that runs on that CPU. 1710 1711 The Linux Performance Event subsystem provides an abstraction of 1712 these software and hardware event capabilities, available via a 1713 system call and used by the "perf" utility in tools/perf/. It 1714 provides per task and per CPU counters, and it provides event 1715 capabilities on top of those. 1716 1717 Say Y if unsure. 1718 1719config DEBUG_PERF_USE_VMALLOC 1720 default n 1721 bool "Debug: use vmalloc to back perf mmap() buffers" 1722 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC 1723 select PERF_USE_VMALLOC 1724 help 1725 Use vmalloc memory to back perf mmap() buffers. 1726 1727 Mostly useful for debugging the vmalloc code on platforms 1728 that don't require it. 1729 1730 Say N if unsure. 1731 1732endmenu 1733 1734config VM_EVENT_COUNTERS 1735 default y 1736 bool "Enable VM event counters for /proc/vmstat" if EXPERT 1737 help 1738 VM event counters are needed for event counts to be shown. 1739 This option allows the disabling of the VM event counters 1740 on EXPERT systems. /proc/vmstat will only show page counts 1741 if VM event counters are disabled. 1742 1743config SLUB_DEBUG 1744 default y 1745 bool "Enable SLUB debugging support" if EXPERT 1746 depends on SLUB && SYSFS 1747 help 1748 SLUB has extensive debug support features. Disabling these can 1749 result in significant savings in code size. This also disables 1750 SLUB sysfs support. /sys/slab will not exist and there will be 1751 no support for cache validation etc. 1752 1753config SLUB_MEMCG_SYSFS_ON 1754 default n 1755 bool "Enable memcg SLUB sysfs support by default" if EXPERT 1756 depends on SLUB && SYSFS && MEMCG 1757 help 1758 SLUB creates a directory under /sys/kernel/slab for each 1759 allocation cache to host info and debug files. If memory 1760 cgroup is enabled, each cache can have per memory cgroup 1761 caches. SLUB can create the same sysfs directories for these 1762 caches under /sys/kernel/slab/CACHE/cgroup but it can lead 1763 to a very high number of debug files being created. This is 1764 controlled by slub_memcg_sysfs boot parameter and this 1765 config option determines the parameter's default value. 1766 1767config COMPAT_BRK 1768 bool "Disable heap randomization" 1769 default y 1770 help 1771 Randomizing heap placement makes heap exploits harder, but it 1772 also breaks ancient binaries (including anything libc5 based). 1773 This option changes the bootup default to heap randomization 1774 disabled, and can be overridden at runtime by setting 1775 /proc/sys/kernel/randomize_va_space to 2. 1776 1777 On non-ancient distros (post-2000 ones) N is usually a safe choice. 1778 1779choice 1780 prompt "Choose SLAB allocator" 1781 default SLUB 1782 help 1783 This option allows to select a slab allocator. 1784 1785config SLAB 1786 bool "SLAB" 1787 select HAVE_HARDENED_USERCOPY_ALLOCATOR 1788 help 1789 The regular slab allocator that is established and known to work 1790 well in all environments. It organizes cache hot objects in 1791 per cpu and per node queues. 1792 1793config SLUB 1794 bool "SLUB (Unqueued Allocator)" 1795 select HAVE_HARDENED_USERCOPY_ALLOCATOR 1796 help 1797 SLUB is a slab allocator that minimizes cache line usage 1798 instead of managing queues of cached objects (SLAB approach). 1799 Per cpu caching is realized using slabs of objects instead 1800 of queues of objects. SLUB can use memory efficiently 1801 and has enhanced diagnostics. SLUB is the default choice for 1802 a slab allocator. 1803 1804config SLOB 1805 depends on EXPERT 1806 bool "SLOB (Simple Allocator)" 1807 help 1808 SLOB replaces the stock allocator with a drastically simpler 1809 allocator. SLOB is generally more space efficient but 1810 does not perform as well on large systems. 1811 1812endchoice 1813 1814config SLAB_MERGE_DEFAULT 1815 bool "Allow slab caches to be merged" 1816 default y 1817 help 1818 For reduced kernel memory fragmentation, slab caches can be 1819 merged when they share the same size and other characteristics. 1820 This carries a risk of kernel heap overflows being able to 1821 overwrite objects from merged caches (and more easily control 1822 cache layout), which makes such heap attacks easier to exploit 1823 by attackers. By keeping caches unmerged, these kinds of exploits 1824 can usually only damage objects in the same cache. To disable 1825 merging at runtime, "slab_nomerge" can be passed on the kernel 1826 command line. 1827 1828config SLAB_FREELIST_RANDOM 1829 default n 1830 depends on SLAB || SLUB 1831 bool "SLAB freelist randomization" 1832 help 1833 Randomizes the freelist order used on creating new pages. This 1834 security feature reduces the predictability of the kernel slab 1835 allocator against heap overflows. 1836 1837config SLAB_FREELIST_HARDENED 1838 bool "Harden slab freelist metadata" 1839 depends on SLUB 1840 help 1841 Many kernel heap attacks try to target slab cache metadata and 1842 other infrastructure. This options makes minor performance 1843 sacrifices to harden the kernel slab allocator against common 1844 freelist exploit methods. 1845 1846config SHUFFLE_PAGE_ALLOCATOR 1847 bool "Page allocator randomization" 1848 default SLAB_FREELIST_RANDOM && ACPI_NUMA 1849 help 1850 Randomization of the page allocator improves the average 1851 utilization of a direct-mapped memory-side-cache. See section 1852 5.2.27 Heterogeneous Memory Attribute Table (HMAT) in the ACPI 1853 6.2a specification for an example of how a platform advertises 1854 the presence of a memory-side-cache. There are also incidental 1855 security benefits as it reduces the predictability of page 1856 allocations to compliment SLAB_FREELIST_RANDOM, but the 1857 default granularity of shuffling on the "MAX_ORDER - 1" i.e, 1858 10th order of pages is selected based on cache utilization 1859 benefits on x86. 1860 1861 While the randomization improves cache utilization it may 1862 negatively impact workloads on platforms without a cache. For 1863 this reason, by default, the randomization is enabled only 1864 after runtime detection of a direct-mapped memory-side-cache. 1865 Otherwise, the randomization may be force enabled with the 1866 'page_alloc.shuffle' kernel command line parameter. 1867 1868 Say Y if unsure. 1869 1870config SLUB_CPU_PARTIAL 1871 default y 1872 depends on SLUB && SMP 1873 bool "SLUB per cpu partial cache" 1874 help 1875 Per cpu partial caches accelerate objects allocation and freeing 1876 that is local to a processor at the price of more indeterminism 1877 in the latency of the free. On overflow these caches will be cleared 1878 which requires the taking of locks that may cause latency spikes. 1879 Typically one would choose no for a realtime system. 1880 1881config MMAP_ALLOW_UNINITIALIZED 1882 bool "Allow mmapped anonymous memory to be uninitialized" 1883 depends on EXPERT && !MMU 1884 default n 1885 help 1886 Normally, and according to the Linux spec, anonymous memory obtained 1887 from mmap() has its contents cleared before it is passed to 1888 userspace. Enabling this config option allows you to request that 1889 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus 1890 providing a huge performance boost. If this option is not enabled, 1891 then the flag will be ignored. 1892 1893 This is taken advantage of by uClibc's malloc(), and also by 1894 ELF-FDPIC binfmt's brk and stack allocator. 1895 1896 Because of the obvious security issues, this option should only be 1897 enabled on embedded devices where you control what is run in 1898 userspace. Since that isn't generally a problem on no-MMU systems, 1899 it is normally safe to say Y here. 1900 1901 See Documentation/nommu-mmap.txt for more information. 1902 1903config SYSTEM_DATA_VERIFICATION 1904 def_bool n 1905 select SYSTEM_TRUSTED_KEYRING 1906 select KEYS 1907 select CRYPTO 1908 select CRYPTO_RSA 1909 select ASYMMETRIC_KEY_TYPE 1910 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE 1911 select ASN1 1912 select OID_REGISTRY 1913 select X509_CERTIFICATE_PARSER 1914 select PKCS7_MESSAGE_PARSER 1915 help 1916 Provide PKCS#7 message verification using the contents of the system 1917 trusted keyring to provide public keys. This then can be used for 1918 module verification, kexec image verification and firmware blob 1919 verification. 1920 1921config PROFILING 1922 bool "Profiling support" 1923 help 1924 Say Y here to enable the extended profiling support mechanisms used 1925 by profilers such as OProfile. 1926 1927# 1928# Place an empty function call at each tracepoint site. Can be 1929# dynamically changed for a probe function. 1930# 1931config TRACEPOINTS 1932 bool 1933 1934endmenu # General setup 1935 1936source "arch/Kconfig" 1937 1938config RT_MUTEXES 1939 bool 1940 1941config BASE_SMALL 1942 int 1943 default 0 if BASE_FULL 1944 default 1 if !BASE_FULL 1945 1946config MODULE_SIG_FORMAT 1947 def_bool n 1948 select SYSTEM_DATA_VERIFICATION 1949 1950menuconfig MODULES 1951 bool "Enable loadable module support" 1952 option modules 1953 help 1954 Kernel modules are small pieces of compiled code which can 1955 be inserted in the running kernel, rather than being 1956 permanently built into the kernel. You use the "modprobe" 1957 tool to add (and sometimes remove) them. If you say Y here, 1958 many parts of the kernel can be built as modules (by 1959 answering M instead of Y where indicated): this is most 1960 useful for infrequently used options which are not required 1961 for booting. For more information, see the man pages for 1962 modprobe, lsmod, modinfo, insmod and rmmod. 1963 1964 If you say Y here, you will need to run "make 1965 modules_install" to put the modules under /lib/modules/ 1966 where modprobe can find them (you may need to be root to do 1967 this). 1968 1969 If unsure, say Y. 1970 1971if MODULES 1972 1973config MODULE_FORCE_LOAD 1974 bool "Forced module loading" 1975 default n 1976 help 1977 Allow loading of modules without version information (ie. modprobe 1978 --force). Forced module loading sets the 'F' (forced) taint flag and 1979 is usually a really bad idea. 1980 1981config MODULE_UNLOAD 1982 bool "Module unloading" 1983 help 1984 Without this option you will not be able to unload any 1985 modules (note that some modules may not be unloadable 1986 anyway), which makes your kernel smaller, faster 1987 and simpler. If unsure, say Y. 1988 1989config MODULE_FORCE_UNLOAD 1990 bool "Forced module unloading" 1991 depends on MODULE_UNLOAD 1992 help 1993 This option allows you to force a module to unload, even if the 1994 kernel believes it is unsafe: the kernel will remove the module 1995 without waiting for anyone to stop using it (using the -f option to 1996 rmmod). This is mainly for kernel developers and desperate users. 1997 If unsure, say N. 1998 1999config MODVERSIONS 2000 bool "Module versioning support" 2001 help 2002 Usually, you have to use modules compiled with your kernel. 2003 Saying Y here makes it sometimes possible to use modules 2004 compiled for different kernels, by adding enough information 2005 to the modules to (hopefully) spot any changes which would 2006 make them incompatible with the kernel you are running. If 2007 unsure, say N. 2008 2009config ASM_MODVERSIONS 2010 bool 2011 default HAVE_ASM_MODVERSIONS && MODVERSIONS 2012 help 2013 This enables module versioning for exported symbols also from 2014 assembly. This can be enabled only when the target architecture 2015 supports it. 2016 2017config MODULE_REL_CRCS 2018 bool 2019 depends on MODVERSIONS 2020 2021config MODULE_SRCVERSION_ALL 2022 bool "Source checksum for all modules" 2023 help 2024 Modules which contain a MODULE_VERSION get an extra "srcversion" 2025 field inserted into their modinfo section, which contains a 2026 sum of the source files which made it. This helps maintainers 2027 see exactly which source was used to build a module (since 2028 others sometimes change the module source without updating 2029 the version). With this option, such a "srcversion" field 2030 will be created for all modules. If unsure, say N. 2031 2032config MODULE_SIG 2033 bool "Module signature verification" 2034 select MODULE_SIG_FORMAT 2035 help 2036 Check modules for valid signatures upon load: the signature 2037 is simply appended to the module. For more information see 2038 <file:Documentation/admin-guide/module-signing.rst>. 2039 2040 Note that this option adds the OpenSSL development packages as a 2041 kernel build dependency so that the signing tool can use its crypto 2042 library. 2043 2044 You should enable this option if you wish to use either 2045 CONFIG_SECURITY_LOCKDOWN_LSM or lockdown functionality imposed via 2046 another LSM - otherwise unsigned modules will be loadable regardless 2047 of the lockdown policy. 2048 2049 !!!WARNING!!! If you enable this option, you MUST make sure that the 2050 module DOES NOT get stripped after being signed. This includes the 2051 debuginfo strip done by some packagers (such as rpmbuild) and 2052 inclusion into an initramfs that wants the module size reduced. 2053 2054config MODULE_SIG_FORCE 2055 bool "Require modules to be validly signed" 2056 depends on MODULE_SIG 2057 help 2058 Reject unsigned modules or signed modules for which we don't have a 2059 key. Without this, such modules will simply taint the kernel. 2060 2061config MODULE_SIG_ALL 2062 bool "Automatically sign all modules" 2063 default y 2064 depends on MODULE_SIG 2065 help 2066 Sign all modules during make modules_install. Without this option, 2067 modules must be signed manually, using the scripts/sign-file tool. 2068 2069comment "Do not forget to sign required modules with scripts/sign-file" 2070 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL 2071 2072choice 2073 prompt "Which hash algorithm should modules be signed with?" 2074 depends on MODULE_SIG 2075 help 2076 This determines which sort of hashing algorithm will be used during 2077 signature generation. This algorithm _must_ be built into the kernel 2078 directly so that signature verification can take place. It is not 2079 possible to load a signed module containing the algorithm to check 2080 the signature on that module. 2081 2082config MODULE_SIG_SHA1 2083 bool "Sign modules with SHA-1" 2084 select CRYPTO_SHA1 2085 2086config MODULE_SIG_SHA224 2087 bool "Sign modules with SHA-224" 2088 select CRYPTO_SHA256 2089 2090config MODULE_SIG_SHA256 2091 bool "Sign modules with SHA-256" 2092 select CRYPTO_SHA256 2093 2094config MODULE_SIG_SHA384 2095 bool "Sign modules with SHA-384" 2096 select CRYPTO_SHA512 2097 2098config MODULE_SIG_SHA512 2099 bool "Sign modules with SHA-512" 2100 select CRYPTO_SHA512 2101 2102endchoice 2103 2104config MODULE_SIG_HASH 2105 string 2106 depends on MODULE_SIG 2107 default "sha1" if MODULE_SIG_SHA1 2108 default "sha224" if MODULE_SIG_SHA224 2109 default "sha256" if MODULE_SIG_SHA256 2110 default "sha384" if MODULE_SIG_SHA384 2111 default "sha512" if MODULE_SIG_SHA512 2112 2113config MODULE_COMPRESS 2114 bool "Compress modules on installation" 2115 help 2116 2117 Compresses kernel modules when 'make modules_install' is run; gzip or 2118 xz depending on "Compression algorithm" below. 2119 2120 module-init-tools MAY support gzip, and kmod MAY support gzip and xz. 2121 2122 Out-of-tree kernel modules installed using Kbuild will also be 2123 compressed upon installation. 2124 2125 Note: for modules inside an initrd or initramfs, it's more efficient 2126 to compress the whole initrd or initramfs instead. 2127 2128 Note: This is fully compatible with signed modules. 2129 2130 If in doubt, say N. 2131 2132choice 2133 prompt "Compression algorithm" 2134 depends on MODULE_COMPRESS 2135 default MODULE_COMPRESS_GZIP 2136 help 2137 This determines which sort of compression will be used during 2138 'make modules_install'. 2139 2140 GZIP (default) and XZ are supported. 2141 2142config MODULE_COMPRESS_GZIP 2143 bool "GZIP" 2144 2145config MODULE_COMPRESS_XZ 2146 bool "XZ" 2147 2148endchoice 2149 2150config MODULE_ALLOW_MISSING_NAMESPACE_IMPORTS 2151 bool "Allow loading of modules with missing namespace imports" 2152 help 2153 Symbols exported with EXPORT_SYMBOL_NS*() are considered exported in 2154 a namespace. A module that makes use of a symbol exported with such a 2155 namespace is required to import the namespace via MODULE_IMPORT_NS(). 2156 There is no technical reason to enforce correct namespace imports, 2157 but it creates consistency between symbols defining namespaces and 2158 users importing namespaces they make use of. This option relaxes this 2159 requirement and lifts the enforcement when loading a module. 2160 2161 If unsure, say N. 2162 2163config UNUSED_SYMBOLS 2164 bool "Enable unused/obsolete exported symbols" 2165 default y if X86 2166 help 2167 Unused but exported symbols make the kernel needlessly bigger. For 2168 that reason most of these unused exports will soon be removed. This 2169 option is provided temporarily to provide a transition period in case 2170 some external kernel module needs one of these symbols anyway. If you 2171 encounter such a case in your module, consider if you are actually 2172 using the right API. (rationale: since nobody in the kernel is using 2173 this in a module, there is a pretty good chance it's actually the 2174 wrong interface to use). If you really need the symbol, please send a 2175 mail to the linux kernel mailing list mentioning the symbol and why 2176 you really need it, and what the merge plan to the mainline kernel for 2177 your module is. 2178 2179config TRIM_UNUSED_KSYMS 2180 bool "Trim unused exported kernel symbols" 2181 depends on !UNUSED_SYMBOLS 2182 help 2183 The kernel and some modules make many symbols available for 2184 other modules to use via EXPORT_SYMBOL() and variants. Depending 2185 on the set of modules being selected in your kernel configuration, 2186 many of those exported symbols might never be used. 2187 2188 This option allows for unused exported symbols to be dropped from 2189 the build. In turn, this provides the compiler more opportunities 2190 (especially when using LTO) for optimizing the code and reducing 2191 binary size. This might have some security advantages as well. 2192 2193 If unsure, or if you need to build out-of-tree modules, say N. 2194 2195endif # MODULES 2196 2197config MODULES_TREE_LOOKUP 2198 def_bool y 2199 depends on PERF_EVENTS || TRACING 2200 2201config INIT_ALL_POSSIBLE 2202 bool 2203 help 2204 Back when each arch used to define their own cpu_online_mask and 2205 cpu_possible_mask, some of them chose to initialize cpu_possible_mask 2206 with all 1s, and others with all 0s. When they were centralised, 2207 it was better to provide this option than to break all the archs 2208 and have several arch maintainers pursuing me down dark alleys. 2209 2210source "block/Kconfig" 2211 2212config PREEMPT_NOTIFIERS 2213 bool 2214 2215config PADATA 2216 depends on SMP 2217 bool 2218 2219config ASN1 2220 tristate 2221 help 2222 Build a simple ASN.1 grammar compiler that produces a bytecode output 2223 that can be interpreted by the ASN.1 stream decoder and used to 2224 inform it as to what tags are to be expected in a stream and what 2225 functions to call on what tags. 2226 2227source "kernel/Kconfig.locks" 2228 2229config ARCH_HAS_SYNC_CORE_BEFORE_USERMODE 2230 bool 2231 2232# It may be useful for an architecture to override the definitions of the 2233# SYSCALL_DEFINE() and __SYSCALL_DEFINEx() macros in <linux/syscalls.h> 2234# and the COMPAT_ variants in <linux/compat.h>, in particular to use a 2235# different calling convention for syscalls. They can also override the 2236# macros for not-implemented syscalls in kernel/sys_ni.c and 2237# kernel/time/posix-stubs.c. All these overrides need to be available in 2238# <asm/syscall_wrapper.h>. 2239config ARCH_HAS_SYSCALL_WRAPPER 2240 def_bool n 2241