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