1config ARCH 2 string 3 option env="ARCH" 4 5config KERNELVERSION 6 string 7 option env="KERNELVERSION" 8 9config DEFCONFIG_LIST 10 string 11 depends on !UML 12 option defconfig_list 13 default "/lib/modules/$UNAME_RELEASE/.config" 14 default "/etc/kernel-config" 15 default "/boot/config-$UNAME_RELEASE" 16 default "$ARCH_DEFCONFIG" 17 default "arch/$ARCH/defconfig" 18 19config CONSTRUCTORS 20 bool 21 depends on !UML 22 23config HAVE_IRQ_WORK 24 bool 25 26config IRQ_WORK 27 bool 28 depends on HAVE_IRQ_WORK 29 30config BUILDTIME_EXTABLE_SORT 31 bool 32 33menu "General setup" 34 35config EXPERIMENTAL 36 bool "Prompt for development and/or incomplete code/drivers" 37 ---help--- 38 Some of the various things that Linux supports (such as network 39 drivers, file systems, network protocols, etc.) can be in a state 40 of development where the functionality, stability, or the level of 41 testing is not yet high enough for general use. This is usually 42 known as the "alpha-test" phase among developers. If a feature is 43 currently in alpha-test, then the developers usually discourage 44 uninformed widespread use of this feature by the general public to 45 avoid "Why doesn't this work?" type mail messages. However, active 46 testing and use of these systems is welcomed. Just be aware that it 47 may not meet the normal level of reliability or it may fail to work 48 in some special cases. Detailed bug reports from people familiar 49 with the kernel internals are usually welcomed by the developers 50 (before submitting bug reports, please read the documents 51 <file:README>, <file:MAINTAINERS>, <file:REPORTING-BUGS>, 52 <file:Documentation/BUG-HUNTING>, and 53 <file:Documentation/oops-tracing.txt> in the kernel source). 54 55 This option will also make obsoleted drivers available. These are 56 drivers that have been replaced by something else, and/or are 57 scheduled to be removed in a future kernel release. 58 59 Unless you intend to help test and develop a feature or driver that 60 falls into this category, or you have a situation that requires 61 using these features, you should probably say N here, which will 62 cause the configurator to present you with fewer choices. If 63 you say Y here, you will be offered the choice of using features or 64 drivers that are currently considered to be in the alpha-test phase. 65 66config BROKEN 67 bool 68 69config BROKEN_ON_SMP 70 bool 71 depends on BROKEN || !SMP 72 default y 73 74config INIT_ENV_ARG_LIMIT 75 int 76 default 32 if !UML 77 default 128 if UML 78 help 79 Maximum of each of the number of arguments and environment 80 variables passed to init from the kernel command line. 81 82 83config CROSS_COMPILE 84 string "Cross-compiler tool prefix" 85 help 86 Same as running 'make CROSS_COMPILE=prefix-' but stored for 87 default make runs in this kernel build directory. You don't 88 need to set this unless you want the configured kernel build 89 directory to select the cross-compiler automatically. 90 91config LOCALVERSION 92 string "Local version - append to kernel release" 93 help 94 Append an extra string to the end of your kernel version. 95 This will show up when you type uname, for example. 96 The string you set here will be appended after the contents of 97 any files with a filename matching localversion* in your 98 object and source tree, in that order. Your total string can 99 be a maximum of 64 characters. 100 101config LOCALVERSION_AUTO 102 bool "Automatically append version information to the version string" 103 default y 104 help 105 This will try to automatically determine if the current tree is a 106 release tree by looking for git tags that belong to the current 107 top of tree revision. 108 109 A string of the format -gxxxxxxxx will be added to the localversion 110 if a git-based tree is found. The string generated by this will be 111 appended after any matching localversion* files, and after the value 112 set in CONFIG_LOCALVERSION. 113 114 (The actual string used here is the first eight characters produced 115 by running the command: 116 117 $ git rev-parse --verify HEAD 118 119 which is done within the script "scripts/setlocalversion".) 120 121config HAVE_KERNEL_GZIP 122 bool 123 124config HAVE_KERNEL_BZIP2 125 bool 126 127config HAVE_KERNEL_LZMA 128 bool 129 130config HAVE_KERNEL_XZ 131 bool 132 133config HAVE_KERNEL_LZO 134 bool 135 136choice 137 prompt "Kernel compression mode" 138 default KERNEL_GZIP 139 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO 140 help 141 The linux kernel is a kind of self-extracting executable. 142 Several compression algorithms are available, which differ 143 in efficiency, compression and decompression speed. 144 Compression speed is only relevant when building a kernel. 145 Decompression speed is relevant at each boot. 146 147 If you have any problems with bzip2 or lzma compressed 148 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older 149 version of this functionality (bzip2 only), for 2.4, was 150 supplied by Christian Ludwig) 151 152 High compression options are mostly useful for users, who 153 are low on disk space (embedded systems), but for whom ram 154 size matters less. 155 156 If in doubt, select 'gzip' 157 158config KERNEL_GZIP 159 bool "Gzip" 160 depends on HAVE_KERNEL_GZIP 161 help 162 The old and tried gzip compression. It provides a good balance 163 between compression ratio and decompression speed. 164 165config KERNEL_BZIP2 166 bool "Bzip2" 167 depends on HAVE_KERNEL_BZIP2 168 help 169 Its compression ratio and speed is intermediate. 170 Decompression speed is slowest among the choices. The kernel 171 size is about 10% smaller with bzip2, in comparison to gzip. 172 Bzip2 uses a large amount of memory. For modern kernels you 173 will need at least 8MB RAM or more for booting. 174 175config KERNEL_LZMA 176 bool "LZMA" 177 depends on HAVE_KERNEL_LZMA 178 help 179 This compression algorithm's ratio is best. Decompression speed 180 is between gzip and bzip2. Compression is slowest. 181 The kernel size is about 33% smaller with LZMA in comparison to gzip. 182 183config KERNEL_XZ 184 bool "XZ" 185 depends on HAVE_KERNEL_XZ 186 help 187 XZ uses the LZMA2 algorithm and instruction set specific 188 BCJ filters which can improve compression ratio of executable 189 code. The size of the kernel is about 30% smaller with XZ in 190 comparison to gzip. On architectures for which there is a BCJ 191 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ 192 will create a few percent smaller kernel than plain LZMA. 193 194 The speed is about the same as with LZMA: The decompression 195 speed of XZ is better than that of bzip2 but worse than gzip 196 and LZO. Compression is slow. 197 198config KERNEL_LZO 199 bool "LZO" 200 depends on HAVE_KERNEL_LZO 201 help 202 Its compression ratio is the poorest among the choices. The kernel 203 size is about 10% bigger than gzip; however its speed 204 (both compression and decompression) is the fastest. 205 206endchoice 207 208config DEFAULT_HOSTNAME 209 string "Default hostname" 210 default "(none)" 211 help 212 This option determines the default system hostname before userspace 213 calls sethostname(2). The kernel traditionally uses "(none)" here, 214 but you may wish to use a different default here to make a minimal 215 system more usable with less configuration. 216 217config SWAP 218 bool "Support for paging of anonymous memory (swap)" 219 depends on MMU && BLOCK 220 default y 221 help 222 This option allows you to choose whether you want to have support 223 for so called swap devices or swap files in your kernel that are 224 used to provide more virtual memory than the actual RAM present 225 in your computer. If unsure say Y. 226 227config SYSVIPC 228 bool "System V IPC" 229 ---help--- 230 Inter Process Communication is a suite of library functions and 231 system calls which let processes (running programs) synchronize and 232 exchange information. It is generally considered to be a good thing, 233 and some programs won't run unless you say Y here. In particular, if 234 you want to run the DOS emulator dosemu under Linux (read the 235 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>), 236 you'll need to say Y here. 237 238 You can find documentation about IPC with "info ipc" and also in 239 section 6.4 of the Linux Programmer's Guide, available from 240 <http://www.tldp.org/guides.html>. 241 242config SYSVIPC_SYSCTL 243 bool 244 depends on SYSVIPC 245 depends on SYSCTL 246 default y 247 248config POSIX_MQUEUE 249 bool "POSIX Message Queues" 250 depends on NET && EXPERIMENTAL 251 ---help--- 252 POSIX variant of message queues is a part of IPC. In POSIX message 253 queues every message has a priority which decides about succession 254 of receiving it by a process. If you want to compile and run 255 programs written e.g. for Solaris with use of its POSIX message 256 queues (functions mq_*) say Y here. 257 258 POSIX message queues are visible as a filesystem called 'mqueue' 259 and can be mounted somewhere if you want to do filesystem 260 operations on message queues. 261 262 If unsure, say Y. 263 264config POSIX_MQUEUE_SYSCTL 265 bool 266 depends on POSIX_MQUEUE 267 depends on SYSCTL 268 default y 269 270config FHANDLE 271 bool "open by fhandle syscalls" 272 select EXPORTFS 273 help 274 If you say Y here, a user level program will be able to map 275 file names to handle and then later use the handle for 276 different file system operations. This is useful in implementing 277 userspace file servers, which now track files using handles instead 278 of names. The handle would remain the same even if file names 279 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2) 280 syscalls. 281 282config AUDIT 283 bool "Auditing support" 284 depends on NET 285 help 286 Enable auditing infrastructure that can be used with another 287 kernel subsystem, such as SELinux (which requires this for 288 logging of avc messages output). Does not do system-call 289 auditing without CONFIG_AUDITSYSCALL. 290 291config AUDITSYSCALL 292 bool "Enable system-call auditing support" 293 depends on AUDIT && (X86 || PPC || S390 || IA64 || UML || SPARC64 || SUPERH || (ARM && AEABI && !OABI_COMPAT)) 294 default y if SECURITY_SELINUX 295 help 296 Enable low-overhead system-call auditing infrastructure that 297 can be used independently or with another kernel subsystem, 298 such as SELinux. 299 300config AUDIT_WATCH 301 def_bool y 302 depends on AUDITSYSCALL 303 select FSNOTIFY 304 305config AUDIT_TREE 306 def_bool y 307 depends on AUDITSYSCALL 308 select FSNOTIFY 309 310config AUDIT_LOGINUID_IMMUTABLE 311 bool "Make audit loginuid immutable" 312 depends on AUDIT 313 help 314 The config option toggles if a task setting its loginuid requires 315 CAP_SYS_AUDITCONTROL or if that task should require no special permissions 316 but should instead only allow setting its loginuid if it was never 317 previously set. On systems which use systemd or a similar central 318 process to restart login services this should be set to true. On older 319 systems in which an admin would typically have to directly stop and 320 start processes this should be set to false. Setting this to true allows 321 one to drop potentially dangerous capabilites from the login tasks, 322 but may not be backwards compatible with older init systems. 323 324source "kernel/irq/Kconfig" 325source "kernel/time/Kconfig" 326 327menu "CPU/Task time and stats accounting" 328 329choice 330 prompt "Cputime accounting" 331 default TICK_CPU_ACCOUNTING if !PPC64 332 default VIRT_CPU_ACCOUNTING if PPC64 333 334# Kind of a stub config for the pure tick based cputime accounting 335config TICK_CPU_ACCOUNTING 336 bool "Simple tick based cputime accounting" 337 depends on !S390 338 help 339 This is the basic tick based cputime accounting that maintains 340 statistics about user, system and idle time spent on per jiffies 341 granularity. 342 343 If unsure, say Y. 344 345config VIRT_CPU_ACCOUNTING 346 bool "Deterministic task and CPU time accounting" 347 depends on HAVE_VIRT_CPU_ACCOUNTING 348 help 349 Select this option to enable more accurate task and CPU time 350 accounting. This is done by reading a CPU counter on each 351 kernel entry and exit and on transitions within the kernel 352 between system, softirq and hardirq state, so there is a 353 small performance impact. In the case of s390 or IBM POWER > 5, 354 this also enables accounting of stolen time on logically-partitioned 355 systems. 356 357config IRQ_TIME_ACCOUNTING 358 bool "Fine granularity task level IRQ time accounting" 359 depends on HAVE_IRQ_TIME_ACCOUNTING 360 help 361 Select this option to enable fine granularity task irq time 362 accounting. This is done by reading a timestamp on each 363 transitions between softirq and hardirq state, so there can be a 364 small performance impact. 365 366 If in doubt, say N here. 367 368endchoice 369 370config BSD_PROCESS_ACCT 371 bool "BSD Process Accounting" 372 help 373 If you say Y here, a user level program will be able to instruct the 374 kernel (via a special system call) to write process accounting 375 information to a file: whenever a process exits, information about 376 that process will be appended to the file by the kernel. The 377 information includes things such as creation time, owning user, 378 command name, memory usage, controlling terminal etc. (the complete 379 list is in the struct acct in <file:include/linux/acct.h>). It is 380 up to the user level program to do useful things with this 381 information. This is generally a good idea, so say Y. 382 383config BSD_PROCESS_ACCT_V3 384 bool "BSD Process Accounting version 3 file format" 385 depends on BSD_PROCESS_ACCT 386 default n 387 help 388 If you say Y here, the process accounting information is written 389 in a new file format that also logs the process IDs of each 390 process and it's parent. Note that this file format is incompatible 391 with previous v0/v1/v2 file formats, so you will need updated tools 392 for processing it. A preliminary version of these tools is available 393 at <http://www.gnu.org/software/acct/>. 394 395config TASKSTATS 396 bool "Export task/process statistics through netlink (EXPERIMENTAL)" 397 depends on NET 398 default n 399 help 400 Export selected statistics for tasks/processes through the 401 generic netlink interface. Unlike BSD process accounting, the 402 statistics are available during the lifetime of tasks/processes as 403 responses to commands. Like BSD accounting, they are sent to user 404 space on task exit. 405 406 Say N if unsure. 407 408config TASK_DELAY_ACCT 409 bool "Enable per-task delay accounting (EXPERIMENTAL)" 410 depends on TASKSTATS 411 help 412 Collect information on time spent by a task waiting for system 413 resources like cpu, synchronous block I/O completion and swapping 414 in pages. Such statistics can help in setting a task's priorities 415 relative to other tasks for cpu, io, rss limits etc. 416 417 Say N if unsure. 418 419config TASK_XACCT 420 bool "Enable extended accounting over taskstats (EXPERIMENTAL)" 421 depends on TASKSTATS 422 help 423 Collect extended task accounting data and send the data 424 to userland for processing over the taskstats interface. 425 426 Say N if unsure. 427 428config TASK_IO_ACCOUNTING 429 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)" 430 depends on TASK_XACCT 431 help 432 Collect information on the number of bytes of storage I/O which this 433 task has caused. 434 435 Say N if unsure. 436 437endmenu # "CPU/Task time and stats accounting" 438 439menu "RCU Subsystem" 440 441choice 442 prompt "RCU Implementation" 443 default TREE_RCU 444 445config TREE_RCU 446 bool "Tree-based hierarchical RCU" 447 depends on !PREEMPT && SMP 448 help 449 This option selects the RCU implementation that is 450 designed for very large SMP system with hundreds or 451 thousands of CPUs. It also scales down nicely to 452 smaller systems. 453 454config TREE_PREEMPT_RCU 455 bool "Preemptible tree-based hierarchical RCU" 456 depends on PREEMPT && SMP 457 help 458 This option selects the RCU implementation that is 459 designed for very large SMP systems with hundreds or 460 thousands of CPUs, but for which real-time response 461 is also required. It also scales down nicely to 462 smaller systems. 463 464config TINY_RCU 465 bool "UP-only small-memory-footprint RCU" 466 depends on !PREEMPT && !SMP 467 help 468 This option selects the RCU implementation that is 469 designed for UP systems from which real-time response 470 is not required. This option greatly reduces the 471 memory footprint of RCU. 472 473config TINY_PREEMPT_RCU 474 bool "Preemptible UP-only small-memory-footprint RCU" 475 depends on PREEMPT && !SMP 476 help 477 This option selects the RCU implementation that is designed 478 for real-time UP systems. This option greatly reduces the 479 memory footprint of RCU. 480 481endchoice 482 483config PREEMPT_RCU 484 def_bool ( TREE_PREEMPT_RCU || TINY_PREEMPT_RCU ) 485 help 486 This option enables preemptible-RCU code that is common between 487 the TREE_PREEMPT_RCU and TINY_PREEMPT_RCU implementations. 488 489config RCU_USER_QS 490 bool "Consider userspace as in RCU extended quiescent state" 491 depends on HAVE_RCU_USER_QS && SMP 492 help 493 This option sets hooks on kernel / userspace boundaries and 494 puts RCU in extended quiescent state when the CPU runs in 495 userspace. It means that when a CPU runs in userspace, it is 496 excluded from the global RCU state machine and thus doesn't 497 to keep the timer tick on for RCU. 498 499config RCU_USER_QS_FORCE 500 bool "Force userspace extended QS by default" 501 depends on RCU_USER_QS 502 help 503 Set the hooks in user/kernel boundaries by default in order to 504 test this feature that treats userspace as an extended quiescent 505 state until we have a real user like a full adaptive nohz option. 506 507config RCU_FANOUT 508 int "Tree-based hierarchical RCU fanout value" 509 range 2 64 if 64BIT 510 range 2 32 if !64BIT 511 depends on TREE_RCU || TREE_PREEMPT_RCU 512 default 64 if 64BIT 513 default 32 if !64BIT 514 help 515 This option controls the fanout of hierarchical implementations 516 of RCU, allowing RCU to work efficiently on machines with 517 large numbers of CPUs. This value must be at least the fourth 518 root of NR_CPUS, which allows NR_CPUS to be insanely large. 519 The default value of RCU_FANOUT should be used for production 520 systems, but if you are stress-testing the RCU implementation 521 itself, small RCU_FANOUT values allow you to test large-system 522 code paths on small(er) systems. 523 524 Select a specific number if testing RCU itself. 525 Take the default if unsure. 526 527config RCU_FANOUT_LEAF 528 int "Tree-based hierarchical RCU leaf-level fanout value" 529 range 2 RCU_FANOUT if 64BIT 530 range 2 RCU_FANOUT if !64BIT 531 depends on TREE_RCU || TREE_PREEMPT_RCU 532 default 16 533 help 534 This option controls the leaf-level fanout of hierarchical 535 implementations of RCU, and allows trading off cache misses 536 against lock contention. Systems that synchronize their 537 scheduling-clock interrupts for energy-efficiency reasons will 538 want the default because the smaller leaf-level fanout keeps 539 lock contention levels acceptably low. Very large systems 540 (hundreds or thousands of CPUs) will instead want to set this 541 value to the maximum value possible in order to reduce the 542 number of cache misses incurred during RCU's grace-period 543 initialization. These systems tend to run CPU-bound, and thus 544 are not helped by synchronized interrupts, and thus tend to 545 skew them, which reduces lock contention enough that large 546 leaf-level fanouts work well. 547 548 Select a specific number if testing RCU itself. 549 550 Select the maximum permissible value for large systems. 551 552 Take the default if unsure. 553 554config RCU_FANOUT_EXACT 555 bool "Disable tree-based hierarchical RCU auto-balancing" 556 depends on TREE_RCU || TREE_PREEMPT_RCU 557 default n 558 help 559 This option forces use of the exact RCU_FANOUT value specified, 560 regardless of imbalances in the hierarchy. This is useful for 561 testing RCU itself, and might one day be useful on systems with 562 strong NUMA behavior. 563 564 Without RCU_FANOUT_EXACT, the code will balance the hierarchy. 565 566 Say N if unsure. 567 568config RCU_FAST_NO_HZ 569 bool "Accelerate last non-dyntick-idle CPU's grace periods" 570 depends on NO_HZ && SMP 571 default n 572 help 573 This option causes RCU to attempt to accelerate grace periods 574 in order to allow CPUs to enter dynticks-idle state more 575 quickly. On the other hand, this option increases the overhead 576 of the dynticks-idle checking, particularly on systems with 577 large numbers of CPUs. 578 579 Say Y if energy efficiency is critically important, particularly 580 if you have relatively few CPUs. 581 582 Say N if you are unsure. 583 584config TREE_RCU_TRACE 585 def_bool RCU_TRACE && ( TREE_RCU || TREE_PREEMPT_RCU ) 586 select DEBUG_FS 587 help 588 This option provides tracing for the TREE_RCU and 589 TREE_PREEMPT_RCU implementations, permitting Makefile to 590 trivially select kernel/rcutree_trace.c. 591 592config RCU_BOOST 593 bool "Enable RCU priority boosting" 594 depends on RT_MUTEXES && PREEMPT_RCU 595 default n 596 help 597 This option boosts the priority of preempted RCU readers that 598 block the current preemptible RCU grace period for too long. 599 This option also prevents heavy loads from blocking RCU 600 callback invocation for all flavors of RCU. 601 602 Say Y here if you are working with real-time apps or heavy loads 603 Say N here if you are unsure. 604 605config RCU_BOOST_PRIO 606 int "Real-time priority to boost RCU readers to" 607 range 1 99 608 depends on RCU_BOOST 609 default 1 610 help 611 This option specifies the real-time priority to which long-term 612 preempted RCU readers are to be boosted. If you are working 613 with a real-time application that has one or more CPU-bound 614 threads running at a real-time priority level, you should set 615 RCU_BOOST_PRIO to a priority higher then the highest-priority 616 real-time CPU-bound thread. The default RCU_BOOST_PRIO value 617 of 1 is appropriate in the common case, which is real-time 618 applications that do not have any CPU-bound threads. 619 620 Some real-time applications might not have a single real-time 621 thread that saturates a given CPU, but instead might have 622 multiple real-time threads that, taken together, fully utilize 623 that CPU. In this case, you should set RCU_BOOST_PRIO to 624 a priority higher than the lowest-priority thread that is 625 conspiring to prevent the CPU from running any non-real-time 626 tasks. For example, if one thread at priority 10 and another 627 thread at priority 5 are between themselves fully consuming 628 the CPU time on a given CPU, then RCU_BOOST_PRIO should be 629 set to priority 6 or higher. 630 631 Specify the real-time priority, or take the default if unsure. 632 633config RCU_BOOST_DELAY 634 int "Milliseconds to delay boosting after RCU grace-period start" 635 range 0 3000 636 depends on RCU_BOOST 637 default 500 638 help 639 This option specifies the time to wait after the beginning of 640 a given grace period before priority-boosting preempted RCU 641 readers blocking that grace period. Note that any RCU reader 642 blocking an expedited RCU grace period is boosted immediately. 643 644 Accept the default if unsure. 645 646endmenu # "RCU Subsystem" 647 648config IKCONFIG 649 tristate "Kernel .config support" 650 ---help--- 651 This option enables the complete Linux kernel ".config" file 652 contents to be saved in the kernel. It provides documentation 653 of which kernel options are used in a running kernel or in an 654 on-disk kernel. This information can be extracted from the kernel 655 image file with the script scripts/extract-ikconfig and used as 656 input to rebuild the current kernel or to build another kernel. 657 It can also be extracted from a running kernel by reading 658 /proc/config.gz if enabled (below). 659 660config IKCONFIG_PROC 661 bool "Enable access to .config through /proc/config.gz" 662 depends on IKCONFIG && PROC_FS 663 ---help--- 664 This option enables access to the kernel configuration file 665 through /proc/config.gz. 666 667config LOG_BUF_SHIFT 668 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" 669 range 12 21 670 default 17 671 help 672 Select kernel log buffer size as a power of 2. 673 Examples: 674 17 => 128 KB 675 16 => 64 KB 676 15 => 32 KB 677 14 => 16 KB 678 13 => 8 KB 679 12 => 4 KB 680 681# 682# Architectures with an unreliable sched_clock() should select this: 683# 684config HAVE_UNSTABLE_SCHED_CLOCK 685 bool 686 687menuconfig CGROUPS 688 boolean "Control Group support" 689 depends on EVENTFD 690 help 691 This option adds support for grouping sets of processes together, for 692 use with process control subsystems such as Cpusets, CFS, memory 693 controls or device isolation. 694 See 695 - Documentation/scheduler/sched-design-CFS.txt (CFS) 696 - Documentation/cgroups/ (features for grouping, isolation 697 and resource control) 698 699 Say N if unsure. 700 701if CGROUPS 702 703config CGROUP_DEBUG 704 bool "Example debug cgroup subsystem" 705 default n 706 help 707 This option enables a simple cgroup subsystem that 708 exports useful debugging information about the cgroups 709 framework. 710 711 Say N if unsure. 712 713config CGROUP_FREEZER 714 bool "Freezer cgroup subsystem" 715 help 716 Provides a way to freeze and unfreeze all tasks in a 717 cgroup. 718 719config CGROUP_DEVICE 720 bool "Device controller for cgroups" 721 help 722 Provides a cgroup implementing whitelists for devices which 723 a process in the cgroup can mknod or open. 724 725config CPUSETS 726 bool "Cpuset support" 727 help 728 This option will let you create and manage CPUSETs which 729 allow dynamically partitioning a system into sets of CPUs and 730 Memory Nodes and assigning tasks to run only within those sets. 731 This is primarily useful on large SMP or NUMA systems. 732 733 Say N if unsure. 734 735config PROC_PID_CPUSET 736 bool "Include legacy /proc/<pid>/cpuset file" 737 depends on CPUSETS 738 default y 739 740config CGROUP_CPUACCT 741 bool "Simple CPU accounting cgroup subsystem" 742 help 743 Provides a simple Resource Controller for monitoring the 744 total CPU consumed by the tasks in a cgroup. 745 746config RESOURCE_COUNTERS 747 bool "Resource counters" 748 help 749 This option enables controller independent resource accounting 750 infrastructure that works with cgroups. 751 752config MEMCG 753 bool "Memory Resource Controller for Control Groups" 754 depends on RESOURCE_COUNTERS 755 select MM_OWNER 756 help 757 Provides a memory resource controller that manages both anonymous 758 memory and page cache. (See Documentation/cgroups/memory.txt) 759 760 Note that setting this option increases fixed memory overhead 761 associated with each page of memory in the system. By this, 762 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory 763 usage tracking struct at boot. Total amount of this is printed out 764 at boot. 765 766 Only enable when you're ok with these trade offs and really 767 sure you need the memory resource controller. Even when you enable 768 this, you can set "cgroup_disable=memory" at your boot option to 769 disable memory resource controller and you can avoid overheads. 770 (and lose benefits of memory resource controller) 771 772 This config option also selects MM_OWNER config option, which 773 could in turn add some fork/exit overhead. 774 775config MEMCG_SWAP 776 bool "Memory Resource Controller Swap Extension" 777 depends on MEMCG && SWAP 778 help 779 Add swap management feature to memory resource controller. When you 780 enable this, you can limit mem+swap usage per cgroup. In other words, 781 when you disable this, memory resource controller has no cares to 782 usage of swap...a process can exhaust all of the swap. This extension 783 is useful when you want to avoid exhaustion swap but this itself 784 adds more overheads and consumes memory for remembering information. 785 Especially if you use 32bit system or small memory system, please 786 be careful about enabling this. When memory resource controller 787 is disabled by boot option, this will be automatically disabled and 788 there will be no overhead from this. Even when you set this config=y, 789 if boot option "swapaccount=0" is set, swap will not be accounted. 790 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page 791 size is 4096bytes, 512k per 1Gbytes of swap. 792config MEMCG_SWAP_ENABLED 793 bool "Memory Resource Controller Swap Extension enabled by default" 794 depends on MEMCG_SWAP 795 default y 796 help 797 Memory Resource Controller Swap Extension comes with its price in 798 a bigger memory consumption. General purpose distribution kernels 799 which want to enable the feature but keep it disabled by default 800 and let the user enable it by swapaccount boot command line 801 parameter should have this option unselected. 802 For those who want to have the feature enabled by default should 803 select this option (if, for some reason, they need to disable it 804 then swapaccount=0 does the trick). 805config MEMCG_KMEM 806 bool "Memory Resource Controller Kernel Memory accounting (EXPERIMENTAL)" 807 depends on MEMCG && EXPERIMENTAL 808 default n 809 help 810 The Kernel Memory extension for Memory Resource Controller can limit 811 the amount of memory used by kernel objects in the system. Those are 812 fundamentally different from the entities handled by the standard 813 Memory Controller, which are page-based, and can be swapped. Users of 814 the kmem extension can use it to guarantee that no group of processes 815 will ever exhaust kernel resources alone. 816 817config CGROUP_HUGETLB 818 bool "HugeTLB Resource Controller for Control Groups" 819 depends on RESOURCE_COUNTERS && HUGETLB_PAGE && EXPERIMENTAL 820 default n 821 help 822 Provides a cgroup Resource Controller for HugeTLB pages. 823 When you enable this, you can put a per cgroup limit on HugeTLB usage. 824 The limit is enforced during page fault. Since HugeTLB doesn't 825 support page reclaim, enforcing the limit at page fault time implies 826 that, the application will get SIGBUS signal if it tries to access 827 HugeTLB pages beyond its limit. This requires the application to know 828 beforehand how much HugeTLB pages it would require for its use. The 829 control group is tracked in the third page lru pointer. This means 830 that we cannot use the controller with huge page less than 3 pages. 831 832config CGROUP_PERF 833 bool "Enable perf_event per-cpu per-container group (cgroup) monitoring" 834 depends on PERF_EVENTS && CGROUPS 835 help 836 This option extends the per-cpu mode to restrict monitoring to 837 threads which belong to the cgroup specified and run on the 838 designated cpu. 839 840 Say N if unsure. 841 842menuconfig CGROUP_SCHED 843 bool "Group CPU scheduler" 844 default n 845 help 846 This feature lets CPU scheduler recognize task groups and control CPU 847 bandwidth allocation to such task groups. It uses cgroups to group 848 tasks. 849 850if CGROUP_SCHED 851config FAIR_GROUP_SCHED 852 bool "Group scheduling for SCHED_OTHER" 853 depends on CGROUP_SCHED 854 default CGROUP_SCHED 855 856config CFS_BANDWIDTH 857 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" 858 depends on EXPERIMENTAL 859 depends on FAIR_GROUP_SCHED 860 default n 861 help 862 This option allows users to define CPU bandwidth rates (limits) for 863 tasks running within the fair group scheduler. Groups with no limit 864 set are considered to be unconstrained and will run with no 865 restriction. 866 See tip/Documentation/scheduler/sched-bwc.txt for more information. 867 868config RT_GROUP_SCHED 869 bool "Group scheduling for SCHED_RR/FIFO" 870 depends on EXPERIMENTAL 871 depends on CGROUP_SCHED 872 default n 873 help 874 This feature lets you explicitly allocate real CPU bandwidth 875 to task groups. If enabled, it will also make it impossible to 876 schedule realtime tasks for non-root users until you allocate 877 realtime bandwidth for them. 878 See Documentation/scheduler/sched-rt-group.txt for more information. 879 880endif #CGROUP_SCHED 881 882config BLK_CGROUP 883 bool "Block IO controller" 884 depends on BLOCK 885 default n 886 ---help--- 887 Generic block IO controller cgroup interface. This is the common 888 cgroup interface which should be used by various IO controlling 889 policies. 890 891 Currently, CFQ IO scheduler uses it to recognize task groups and 892 control disk bandwidth allocation (proportional time slice allocation) 893 to such task groups. It is also used by bio throttling logic in 894 block layer to implement upper limit in IO rates on a device. 895 896 This option only enables generic Block IO controller infrastructure. 897 One needs to also enable actual IO controlling logic/policy. For 898 enabling proportional weight division of disk bandwidth in CFQ, set 899 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set 900 CONFIG_BLK_DEV_THROTTLING=y. 901 902 See Documentation/cgroups/blkio-controller.txt for more information. 903 904config DEBUG_BLK_CGROUP 905 bool "Enable Block IO controller debugging" 906 depends on BLK_CGROUP 907 default n 908 ---help--- 909 Enable some debugging help. Currently it exports additional stat 910 files in a cgroup which can be useful for debugging. 911 912endif # CGROUPS 913 914config CHECKPOINT_RESTORE 915 bool "Checkpoint/restore support" if EXPERT 916 default n 917 help 918 Enables additional kernel features in a sake of checkpoint/restore. 919 In particular it adds auxiliary prctl codes to setup process text, 920 data and heap segment sizes, and a few additional /proc filesystem 921 entries. 922 923 If unsure, say N here. 924 925menuconfig NAMESPACES 926 bool "Namespaces support" if EXPERT 927 default !EXPERT 928 help 929 Provides the way to make tasks work with different objects using 930 the same id. For example same IPC id may refer to different objects 931 or same user id or pid may refer to different tasks when used in 932 different namespaces. 933 934if NAMESPACES 935 936config UTS_NS 937 bool "UTS namespace" 938 default y 939 help 940 In this namespace tasks see different info provided with the 941 uname() system call 942 943config IPC_NS 944 bool "IPC namespace" 945 depends on (SYSVIPC || POSIX_MQUEUE) 946 default y 947 help 948 In this namespace tasks work with IPC ids which correspond to 949 different IPC objects in different namespaces. 950 951config USER_NS 952 bool "User namespace (EXPERIMENTAL)" 953 depends on EXPERIMENTAL 954 depends on UIDGID_CONVERTED 955 select UIDGID_STRICT_TYPE_CHECKS 956 957 default n 958 help 959 This allows containers, i.e. vservers, to use user namespaces 960 to provide different user info for different servers. 961 If unsure, say N. 962 963config PID_NS 964 bool "PID Namespaces" 965 default y 966 help 967 Support process id namespaces. This allows having multiple 968 processes with the same pid as long as they are in different 969 pid namespaces. This is a building block of containers. 970 971config NET_NS 972 bool "Network namespace" 973 depends on NET 974 default y 975 help 976 Allow user space to create what appear to be multiple instances 977 of the network stack. 978 979endif # NAMESPACES 980 981config UIDGID_CONVERTED 982 # True if all of the selected software conmponents are known 983 # to have uid_t and gid_t converted to kuid_t and kgid_t 984 # where appropriate and are otherwise safe to use with 985 # the user namespace. 986 bool 987 default y 988 989 # List of kernel pieces that need user namespace work 990 # Features 991 depends on SYSVIPC = n 992 depends on IMA = n 993 depends on EVM = n 994 depends on KEYS = n 995 depends on AUDIT = n 996 depends on AUDITSYSCALL = n 997 depends on TASKSTATS = n 998 depends on TRACING = n 999 depends on FS_POSIX_ACL = n 1000 depends on QUOTA = n 1001 depends on QUOTACTL = n 1002 depends on DEBUG_CREDENTIALS = n 1003 depends on BSD_PROCESS_ACCT = n 1004 depends on DRM = n 1005 depends on PROC_EVENTS = n 1006 1007 # Networking 1008 depends on NET = n 1009 depends on NET_9P = n 1010 depends on IPX = n 1011 depends on PHONET = n 1012 depends on NET_CLS_FLOW = n 1013 depends on NETFILTER_XT_MATCH_OWNER = n 1014 depends on NETFILTER_XT_MATCH_RECENT = n 1015 depends on NETFILTER_XT_TARGET_LOG = n 1016 depends on NETFILTER_NETLINK_LOG = n 1017 depends on INET = n 1018 depends on IPV6 = n 1019 depends on IP_SCTP = n 1020 depends on AF_RXRPC = n 1021 depends on LLC2 = n 1022 depends on NET_KEY = n 1023 depends on INET_DIAG = n 1024 depends on DNS_RESOLVER = n 1025 depends on AX25 = n 1026 depends on ATALK = n 1027 1028 # Filesystems 1029 depends on USB_DEVICEFS = n 1030 depends on USB_GADGETFS = n 1031 depends on USB_FUNCTIONFS = n 1032 depends on DEVTMPFS = n 1033 depends on XENFS = n 1034 1035 depends on 9P_FS = n 1036 depends on ADFS_FS = n 1037 depends on AFFS_FS = n 1038 depends on AFS_FS = n 1039 depends on AUTOFS4_FS = n 1040 depends on BEFS_FS = n 1041 depends on BFS_FS = n 1042 depends on BTRFS_FS = n 1043 depends on CEPH_FS = n 1044 depends on CIFS = n 1045 depends on CODA_FS = n 1046 depends on CONFIGFS_FS = n 1047 depends on CRAMFS = n 1048 depends on DEBUG_FS = n 1049 depends on ECRYPT_FS = n 1050 depends on EFS_FS = n 1051 depends on EXOFS_FS = n 1052 depends on FAT_FS = n 1053 depends on FUSE_FS = n 1054 depends on GFS2_FS = n 1055 depends on HFS_FS = n 1056 depends on HFSPLUS_FS = n 1057 depends on HPFS_FS = n 1058 depends on HUGETLBFS = n 1059 depends on ISO9660_FS = n 1060 depends on JFFS2_FS = n 1061 depends on JFS_FS = n 1062 depends on LOGFS = n 1063 depends on MINIX_FS = n 1064 depends on NCP_FS = n 1065 depends on NFSD = n 1066 depends on NFS_FS = n 1067 depends on NILFS2_FS = n 1068 depends on NTFS_FS = n 1069 depends on OCFS2_FS = n 1070 depends on OMFS_FS = n 1071 depends on QNX4FS_FS = n 1072 depends on QNX6FS_FS = n 1073 depends on REISERFS_FS = n 1074 depends on SQUASHFS = n 1075 depends on SYSV_FS = n 1076 depends on UBIFS_FS = n 1077 depends on UDF_FS = n 1078 depends on UFS_FS = n 1079 depends on VXFS_FS = n 1080 depends on XFS_FS = n 1081 1082 depends on !UML || HOSTFS = n 1083 1084 # The rare drivers that won't build 1085 depends on AIRO = n 1086 depends on AIRO_CS = n 1087 depends on TUN = n 1088 depends on INFINIBAND_QIB = n 1089 depends on BLK_DEV_LOOP = n 1090 depends on ANDROID_BINDER_IPC = n 1091 1092 # Security modules 1093 depends on SECURITY_TOMOYO = n 1094 depends on SECURITY_APPARMOR = n 1095 1096config UIDGID_STRICT_TYPE_CHECKS 1097 bool "Require conversions between uid/gids and their internal representation" 1098 depends on UIDGID_CONVERTED 1099 default n 1100 help 1101 While the nececessary conversions are being added to all subsystems this option allows 1102 the code to continue to build for unconverted subsystems. 1103 1104 Say Y here if you want the strict type checking enabled 1105 1106config SCHED_AUTOGROUP 1107 bool "Automatic process group scheduling" 1108 select EVENTFD 1109 select CGROUPS 1110 select CGROUP_SCHED 1111 select FAIR_GROUP_SCHED 1112 help 1113 This option optimizes the scheduler for common desktop workloads by 1114 automatically creating and populating task groups. This separation 1115 of workloads isolates aggressive CPU burners (like build jobs) from 1116 desktop applications. Task group autogeneration is currently based 1117 upon task session. 1118 1119config MM_OWNER 1120 bool 1121 1122config SYSFS_DEPRECATED 1123 bool "Enable deprecated sysfs features to support old userspace tools" 1124 depends on SYSFS 1125 default n 1126 help 1127 This option adds code that switches the layout of the "block" class 1128 devices, to not show up in /sys/class/block/, but only in 1129 /sys/block/. 1130 1131 This switch is only active when the sysfs.deprecated=1 boot option is 1132 passed or the SYSFS_DEPRECATED_V2 option is set. 1133 1134 This option allows new kernels to run on old distributions and tools, 1135 which might get confused by /sys/class/block/. Since 2007/2008 all 1136 major distributions and tools handle this just fine. 1137 1138 Recent distributions and userspace tools after 2009/2010 depend on 1139 the existence of /sys/class/block/, and will not work with this 1140 option enabled. 1141 1142 Only if you are using a new kernel on an old distribution, you might 1143 need to say Y here. 1144 1145config SYSFS_DEPRECATED_V2 1146 bool "Enable deprecated sysfs features by default" 1147 default n 1148 depends on SYSFS 1149 depends on SYSFS_DEPRECATED 1150 help 1151 Enable deprecated sysfs by default. 1152 1153 See the CONFIG_SYSFS_DEPRECATED option for more details about this 1154 option. 1155 1156 Only if you are using a new kernel on an old distribution, you might 1157 need to say Y here. Even then, odds are you would not need it 1158 enabled, you can always pass the boot option if absolutely necessary. 1159 1160config RELAY 1161 bool "Kernel->user space relay support (formerly relayfs)" 1162 help 1163 This option enables support for relay interface support in 1164 certain file systems (such as debugfs). 1165 It is designed to provide an efficient mechanism for tools and 1166 facilities to relay large amounts of data from kernel space to 1167 user space. 1168 1169 If unsure, say N. 1170 1171config BLK_DEV_INITRD 1172 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support" 1173 depends on BROKEN || !FRV 1174 help 1175 The initial RAM filesystem is a ramfs which is loaded by the 1176 boot loader (loadlin or lilo) and that is mounted as root 1177 before the normal boot procedure. It is typically used to 1178 load modules needed to mount the "real" root file system, 1179 etc. See <file:Documentation/initrd.txt> for details. 1180 1181 If RAM disk support (BLK_DEV_RAM) is also included, this 1182 also enables initial RAM disk (initrd) support and adds 1183 15 Kbytes (more on some other architectures) to the kernel size. 1184 1185 If unsure say Y. 1186 1187if BLK_DEV_INITRD 1188 1189source "usr/Kconfig" 1190 1191endif 1192 1193config CC_OPTIMIZE_FOR_SIZE 1194 bool "Optimize for size" 1195 help 1196 Enabling this option will pass "-Os" instead of "-O2" to gcc 1197 resulting in a smaller kernel. 1198 1199 If unsure, say Y. 1200 1201config SYSCTL 1202 bool 1203 1204config ANON_INODES 1205 bool 1206 1207menuconfig EXPERT 1208 bool "Configure standard kernel features (expert users)" 1209 # Unhide debug options, to make the on-by-default options visible 1210 select DEBUG_KERNEL 1211 help 1212 This option allows certain base kernel options and settings 1213 to be disabled or tweaked. This is for specialized 1214 environments which can tolerate a "non-standard" kernel. 1215 Only use this if you really know what you are doing. 1216 1217config UID16 1218 bool "Enable 16-bit UID system calls" if EXPERT 1219 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION) \ 1220 || AARCH32_EMULATION 1221 default y 1222 help 1223 This enables the legacy 16-bit UID syscall wrappers. 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 KALLSYMS 1243 bool "Load all symbols for debugging/ksymoops" if EXPERT 1244 default y 1245 help 1246 Say Y here to let the kernel print out symbolic crash information and 1247 symbolic stack backtraces. This increases the size of the kernel 1248 somewhat, as all symbols have to be loaded into the kernel image. 1249 1250config KALLSYMS_ALL 1251 bool "Include all symbols in kallsyms" 1252 depends on DEBUG_KERNEL && KALLSYMS 1253 help 1254 Normally kallsyms only contains the symbols of functions for nicer 1255 OOPS messages and backtraces (i.e., symbols from the text and inittext 1256 sections). This is sufficient for most cases. And only in very rare 1257 cases (e.g., when a debugger is used) all symbols are required (e.g., 1258 names of variables from the data sections, etc). 1259 1260 This option makes sure that all symbols are loaded into the kernel 1261 image (i.e., symbols from all sections) in cost of increased kernel 1262 size (depending on the kernel configuration, it may be 300KiB or 1263 something like this). 1264 1265 Say N unless you really need all symbols. 1266 1267config HOTPLUG 1268 def_bool y 1269 1270config PRINTK 1271 default y 1272 bool "Enable support for printk" if EXPERT 1273 help 1274 This option enables normal printk support. Removing it 1275 eliminates most of the message strings from the kernel image 1276 and makes the kernel more or less silent. As this makes it 1277 very difficult to diagnose system problems, saying N here is 1278 strongly discouraged. 1279 1280config BUG 1281 bool "BUG() support" if EXPERT 1282 default y 1283 help 1284 Disabling this option eliminates support for BUG and WARN, reducing 1285 the size of your kernel image and potentially quietly ignoring 1286 numerous fatal conditions. You should only consider disabling this 1287 option for embedded systems with no facilities for reporting errors. 1288 Just say Y. 1289 1290config ELF_CORE 1291 default y 1292 bool "Enable ELF core dumps" if EXPERT 1293 help 1294 Enable support for generating core dumps. Disabling saves about 4k. 1295 1296 1297config PCSPKR_PLATFORM 1298 bool "Enable PC-Speaker support" if EXPERT 1299 depends on HAVE_PCSPKR_PLATFORM 1300 select I8253_LOCK 1301 default y 1302 help 1303 This option allows to disable the internal PC-Speaker 1304 support, saving some memory. 1305 1306config HAVE_PCSPKR_PLATFORM 1307 bool 1308 1309config BASE_FULL 1310 default y 1311 bool "Enable full-sized data structures for core" if EXPERT 1312 help 1313 Disabling this option reduces the size of miscellaneous core 1314 kernel data structures. This saves memory on small machines, 1315 but may reduce performance. 1316 1317config FUTEX 1318 bool "Enable futex support" if EXPERT 1319 default y 1320 select RT_MUTEXES 1321 help 1322 Disabling this option will cause the kernel to be built without 1323 support for "fast userspace mutexes". The resulting kernel may not 1324 run glibc-based applications correctly. 1325 1326config EPOLL 1327 bool "Enable eventpoll support" if EXPERT 1328 default y 1329 select ANON_INODES 1330 help 1331 Disabling this option will cause the kernel to be built without 1332 support for epoll family of system calls. 1333 1334config SIGNALFD 1335 bool "Enable signalfd() system call" if EXPERT 1336 select ANON_INODES 1337 default y 1338 help 1339 Enable the signalfd() system call that allows to receive signals 1340 on a file descriptor. 1341 1342 If unsure, say Y. 1343 1344config TIMERFD 1345 bool "Enable timerfd() system call" if EXPERT 1346 select ANON_INODES 1347 default y 1348 help 1349 Enable the timerfd() system call that allows to receive timer 1350 events on a file descriptor. 1351 1352 If unsure, say Y. 1353 1354config EVENTFD 1355 bool "Enable eventfd() system call" if EXPERT 1356 select ANON_INODES 1357 default y 1358 help 1359 Enable the eventfd() system call that allows to receive both 1360 kernel notification (ie. KAIO) or userspace notifications. 1361 1362 If unsure, say Y. 1363 1364config SHMEM 1365 bool "Use full shmem filesystem" if EXPERT 1366 default y 1367 depends on MMU 1368 help 1369 The shmem is an internal filesystem used to manage shared memory. 1370 It is backed by swap and manages resource limits. It is also exported 1371 to userspace as tmpfs if TMPFS is enabled. Disabling this 1372 option replaces shmem and tmpfs with the much simpler ramfs code, 1373 which may be appropriate on small systems without swap. 1374 1375config AIO 1376 bool "Enable AIO support" if EXPERT 1377 default y 1378 help 1379 This option enables POSIX asynchronous I/O which may by used 1380 by some high performance threaded applications. Disabling 1381 this option saves about 7k. 1382 1383config EMBEDDED 1384 bool "Embedded system" 1385 select EXPERT 1386 help 1387 This option should be enabled if compiling the kernel for 1388 an embedded system so certain expert options are available 1389 for configuration. 1390 1391config HAVE_PERF_EVENTS 1392 bool 1393 help 1394 See tools/perf/design.txt for details. 1395 1396config PERF_USE_VMALLOC 1397 bool 1398 help 1399 See tools/perf/design.txt for details 1400 1401menu "Kernel Performance Events And Counters" 1402 1403config PERF_EVENTS 1404 bool "Kernel performance events and counters" 1405 default y if PROFILING 1406 depends on HAVE_PERF_EVENTS 1407 select ANON_INODES 1408 select IRQ_WORK 1409 help 1410 Enable kernel support for various performance events provided 1411 by software and hardware. 1412 1413 Software events are supported either built-in or via the 1414 use of generic tracepoints. 1415 1416 Most modern CPUs support performance events via performance 1417 counter registers. These registers count the number of certain 1418 types of hw events: such as instructions executed, cachemisses 1419 suffered, or branches mis-predicted - without slowing down the 1420 kernel or applications. These registers can also trigger interrupts 1421 when a threshold number of events have passed - and can thus be 1422 used to profile the code that runs on that CPU. 1423 1424 The Linux Performance Event subsystem provides an abstraction of 1425 these software and hardware event capabilities, available via a 1426 system call and used by the "perf" utility in tools/perf/. It 1427 provides per task and per CPU counters, and it provides event 1428 capabilities on top of those. 1429 1430 Say Y if unsure. 1431 1432config DEBUG_PERF_USE_VMALLOC 1433 default n 1434 bool "Debug: use vmalloc to back perf mmap() buffers" 1435 depends on PERF_EVENTS && DEBUG_KERNEL 1436 select PERF_USE_VMALLOC 1437 help 1438 Use vmalloc memory to back perf mmap() buffers. 1439 1440 Mostly useful for debugging the vmalloc code on platforms 1441 that don't require it. 1442 1443 Say N if unsure. 1444 1445endmenu 1446 1447config VM_EVENT_COUNTERS 1448 default y 1449 bool "Enable VM event counters for /proc/vmstat" if EXPERT 1450 help 1451 VM event counters are needed for event counts to be shown. 1452 This option allows the disabling of the VM event counters 1453 on EXPERT systems. /proc/vmstat will only show page counts 1454 if VM event counters are disabled. 1455 1456config PCI_QUIRKS 1457 default y 1458 bool "Enable PCI quirk workarounds" if EXPERT 1459 depends on PCI 1460 help 1461 This enables workarounds for various PCI chipset 1462 bugs/quirks. Disable this only if your target machine is 1463 unaffected by PCI quirks. 1464 1465config SLUB_DEBUG 1466 default y 1467 bool "Enable SLUB debugging support" if EXPERT 1468 depends on SLUB && SYSFS 1469 help 1470 SLUB has extensive debug support features. Disabling these can 1471 result in significant savings in code size. This also disables 1472 SLUB sysfs support. /sys/slab will not exist and there will be 1473 no support for cache validation etc. 1474 1475config COMPAT_BRK 1476 bool "Disable heap randomization" 1477 default y 1478 help 1479 Randomizing heap placement makes heap exploits harder, but it 1480 also breaks ancient binaries (including anything libc5 based). 1481 This option changes the bootup default to heap randomization 1482 disabled, and can be overridden at runtime by setting 1483 /proc/sys/kernel/randomize_va_space to 2. 1484 1485 On non-ancient distros (post-2000 ones) N is usually a safe choice. 1486 1487choice 1488 prompt "Choose SLAB allocator" 1489 default SLUB 1490 help 1491 This option allows to select a slab allocator. 1492 1493config SLAB 1494 bool "SLAB" 1495 help 1496 The regular slab allocator that is established and known to work 1497 well in all environments. It organizes cache hot objects in 1498 per cpu and per node queues. 1499 1500config SLUB 1501 bool "SLUB (Unqueued Allocator)" 1502 help 1503 SLUB is a slab allocator that minimizes cache line usage 1504 instead of managing queues of cached objects (SLAB approach). 1505 Per cpu caching is realized using slabs of objects instead 1506 of queues of objects. SLUB can use memory efficiently 1507 and has enhanced diagnostics. SLUB is the default choice for 1508 a slab allocator. 1509 1510config SLOB 1511 depends on EXPERT 1512 bool "SLOB (Simple Allocator)" 1513 help 1514 SLOB replaces the stock allocator with a drastically simpler 1515 allocator. SLOB is generally more space efficient but 1516 does not perform as well on large systems. 1517 1518endchoice 1519 1520config MMAP_ALLOW_UNINITIALIZED 1521 bool "Allow mmapped anonymous memory to be uninitialized" 1522 depends on EXPERT && !MMU 1523 default n 1524 help 1525 Normally, and according to the Linux spec, anonymous memory obtained 1526 from mmap() has it's contents cleared before it is passed to 1527 userspace. Enabling this config option allows you to request that 1528 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus 1529 providing a huge performance boost. If this option is not enabled, 1530 then the flag will be ignored. 1531 1532 This is taken advantage of by uClibc's malloc(), and also by 1533 ELF-FDPIC binfmt's brk and stack allocator. 1534 1535 Because of the obvious security issues, this option should only be 1536 enabled on embedded devices where you control what is run in 1537 userspace. Since that isn't generally a problem on no-MMU systems, 1538 it is normally safe to say Y here. 1539 1540 See Documentation/nommu-mmap.txt for more information. 1541 1542config PROFILING 1543 bool "Profiling support" 1544 help 1545 Say Y here to enable the extended profiling support mechanisms used 1546 by profilers such as OProfile. 1547 1548# 1549# Place an empty function call at each tracepoint site. Can be 1550# dynamically changed for a probe function. 1551# 1552config TRACEPOINTS 1553 bool 1554 1555source "arch/Kconfig" 1556 1557endmenu # General setup 1558 1559config HAVE_GENERIC_DMA_COHERENT 1560 bool 1561 default n 1562 1563config SLABINFO 1564 bool 1565 depends on PROC_FS 1566 depends on SLAB || SLUB_DEBUG 1567 default y 1568 1569config RT_MUTEXES 1570 boolean 1571 1572config BASE_SMALL 1573 int 1574 default 0 if BASE_FULL 1575 default 1 if !BASE_FULL 1576 1577menuconfig MODULES 1578 bool "Enable loadable module support" 1579 help 1580 Kernel modules are small pieces of compiled code which can 1581 be inserted in the running kernel, rather than being 1582 permanently built into the kernel. You use the "modprobe" 1583 tool to add (and sometimes remove) them. If you say Y here, 1584 many parts of the kernel can be built as modules (by 1585 answering M instead of Y where indicated): this is most 1586 useful for infrequently used options which are not required 1587 for booting. For more information, see the man pages for 1588 modprobe, lsmod, modinfo, insmod and rmmod. 1589 1590 If you say Y here, you will need to run "make 1591 modules_install" to put the modules under /lib/modules/ 1592 where modprobe can find them (you may need to be root to do 1593 this). 1594 1595 If unsure, say Y. 1596 1597if MODULES 1598 1599config MODULE_FORCE_LOAD 1600 bool "Forced module loading" 1601 default n 1602 help 1603 Allow loading of modules without version information (ie. modprobe 1604 --force). Forced module loading sets the 'F' (forced) taint flag and 1605 is usually a really bad idea. 1606 1607config MODULE_UNLOAD 1608 bool "Module unloading" 1609 help 1610 Without this option you will not be able to unload any 1611 modules (note that some modules may not be unloadable 1612 anyway), which makes your kernel smaller, faster 1613 and simpler. If unsure, say Y. 1614 1615config MODULE_FORCE_UNLOAD 1616 bool "Forced module unloading" 1617 depends on MODULE_UNLOAD && EXPERIMENTAL 1618 help 1619 This option allows you to force a module to unload, even if the 1620 kernel believes it is unsafe: the kernel will remove the module 1621 without waiting for anyone to stop using it (using the -f option to 1622 rmmod). This is mainly for kernel developers and desperate users. 1623 If unsure, say N. 1624 1625config MODVERSIONS 1626 bool "Module versioning support" 1627 help 1628 Usually, you have to use modules compiled with your kernel. 1629 Saying Y here makes it sometimes possible to use modules 1630 compiled for different kernels, by adding enough information 1631 to the modules to (hopefully) spot any changes which would 1632 make them incompatible with the kernel you are running. If 1633 unsure, say N. 1634 1635config MODULE_SRCVERSION_ALL 1636 bool "Source checksum for all modules" 1637 help 1638 Modules which contain a MODULE_VERSION get an extra "srcversion" 1639 field inserted into their modinfo section, which contains a 1640 sum of the source files which made it. This helps maintainers 1641 see exactly which source was used to build a module (since 1642 others sometimes change the module source without updating 1643 the version). With this option, such a "srcversion" field 1644 will be created for all modules. If unsure, say N. 1645 1646endif # MODULES 1647 1648config INIT_ALL_POSSIBLE 1649 bool 1650 help 1651 Back when each arch used to define their own cpu_online_mask and 1652 cpu_possible_mask, some of them chose to initialize cpu_possible_mask 1653 with all 1s, and others with all 0s. When they were centralised, 1654 it was better to provide this option than to break all the archs 1655 and have several arch maintainers pursuing me down dark alleys. 1656 1657config STOP_MACHINE 1658 bool 1659 default y 1660 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU 1661 help 1662 Need stop_machine() primitive. 1663 1664source "block/Kconfig" 1665 1666config PREEMPT_NOTIFIERS 1667 bool 1668 1669config PADATA 1670 depends on SMP 1671 bool 1672 1673source "kernel/Kconfig.locks" 1674