blob: 56fd93c63c77d7e9f6735af3f4319c1ba17c8e74 [file] [log] [blame]
Roman Zippel80daa562008-01-14 04:51:16 +01001config ARCH
2 string
3 option env="ARCH"
4
5config KERNELVERSION
6 string
7 option env="KERNELVERSION"
8
Roman Zippelface4372006-06-08 22:12:45 -07009config DEFCONFIG_LIST
10 string
Paolo 'Blaisorblade' Giarrussob2670eac2006-10-19 23:28:23 -070011 depends on !UML
Roman Zippelface4372006-06-08 22:12:45 -070012 option defconfig_list
13 default "/lib/modules/$UNAME_RELEASE/.config"
14 default "/etc/kernel-config"
15 default "/boot/config-$UNAME_RELEASE"
Sam Ravnborg73531902008-05-25 23:03:18 +020016 default "$ARCH_DEFCONFIG"
Roman Zippelface4372006-06-08 22:12:45 -070017 default "arch/$ARCH/defconfig"
18
Al Boldiff0cfc62007-07-31 00:39:23 -070019menu "General setup"
Linus Torvalds1da177e2005-04-16 15:20:36 -070020
21config EXPERIMENTAL
22 bool "Prompt for development and/or incomplete code/drivers"
23 ---help---
24 Some of the various things that Linux supports (such as network
25 drivers, file systems, network protocols, etc.) can be in a state
26 of development where the functionality, stability, or the level of
27 testing is not yet high enough for general use. This is usually
28 known as the "alpha-test" phase among developers. If a feature is
29 currently in alpha-test, then the developers usually discourage
30 uninformed widespread use of this feature by the general public to
31 avoid "Why doesn't this work?" type mail messages. However, active
32 testing and use of these systems is welcomed. Just be aware that it
33 may not meet the normal level of reliability or it may fail to work
34 in some special cases. Detailed bug reports from people familiar
35 with the kernel internals are usually welcomed by the developers
36 (before submitting bug reports, please read the documents
37 <file:README>, <file:MAINTAINERS>, <file:REPORTING-BUGS>,
38 <file:Documentation/BUG-HUNTING>, and
39 <file:Documentation/oops-tracing.txt> in the kernel source).
40
41 This option will also make obsoleted drivers available. These are
42 drivers that have been replaced by something else, and/or are
43 scheduled to be removed in a future kernel release.
44
45 Unless you intend to help test and develop a feature or driver that
46 falls into this category, or you have a situation that requires
47 using these features, you should probably say N here, which will
48 cause the configurator to present you with fewer choices. If
49 you say Y here, you will be offered the choice of using features or
50 drivers that are currently considered to be in the alpha-test phase.
51
Linus Torvalds1da177e2005-04-16 15:20:36 -070052config BROKEN
53 bool
Linus Torvalds1da177e2005-04-16 15:20:36 -070054
55config BROKEN_ON_SMP
56 bool
57 depends on BROKEN || !SMP
58 default y
59
60config LOCK_KERNEL
61 bool
62 depends on SMP || PREEMPT
63 default y
64
65config INIT_ENV_ARG_LIMIT
66 int
Adrian Bunkdd673bc2006-06-30 01:55:51 -070067 default 32 if !UML
68 default 128 if UML
Linus Torvalds1da177e2005-04-16 15:20:36 -070069 help
Randy Dunlap34ad92c22005-10-30 15:01:46 -080070 Maximum of each of the number of arguments and environment
71 variables passed to init from the kernel command line.
Linus Torvalds1da177e2005-04-16 15:20:36 -070072
Linus Torvalds1da177e2005-04-16 15:20:36 -070073
74config LOCALVERSION
75 string "Local version - append to kernel release"
76 help
77 Append an extra string to the end of your kernel version.
78 This will show up when you type uname, for example.
79 The string you set here will be appended after the contents of
80 any files with a filename matching localversion* in your
81 object and source tree, in that order. Your total string can
82 be a maximum of 64 characters.
83
Ryan Andersonaaebf432005-07-31 04:57:49 -040084config LOCALVERSION_AUTO
85 bool "Automatically append version information to the version string"
86 default y
87 help
88 This will try to automatically determine if the current tree is a
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020089 release tree by looking for git tags that belong to the current
90 top of tree revision.
Ryan Andersonaaebf432005-07-31 04:57:49 -040091
92 A string of the format -gxxxxxxxx will be added to the localversion
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020093 if a git-based tree is found. The string generated by this will be
Ryan Andersonaaebf432005-07-31 04:57:49 -040094 appended after any matching localversion* files, and after the value
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020095 set in CONFIG_LOCALVERSION.
Ryan Andersonaaebf432005-07-31 04:57:49 -040096
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020097 (The actual string used here is the first eight characters produced
98 by running the command:
99
100 $ git rev-parse --verify HEAD
101
102 which is done within the script "scripts/setlocalversion".)
Ryan Andersonaaebf432005-07-31 04:57:49 -0400103
Linus Torvalds1da177e2005-04-16 15:20:36 -0700104config SWAP
105 bool "Support for paging of anonymous memory (swap)"
David Howells93614012006-09-30 20:45:40 +0200106 depends on MMU && BLOCK
Linus Torvalds1da177e2005-04-16 15:20:36 -0700107 default y
108 help
109 This option allows you to choose whether you want to have support
Jesper Juhl92c35042006-01-15 02:40:08 +0100110 for so called swap devices or swap files in your kernel that are
Linus Torvalds1da177e2005-04-16 15:20:36 -0700111 used to provide more virtual memory than the actual RAM present
112 in your computer. If unsure say Y.
113
114config SYSVIPC
115 bool "System V IPC"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700116 ---help---
117 Inter Process Communication is a suite of library functions and
118 system calls which let processes (running programs) synchronize and
119 exchange information. It is generally considered to be a good thing,
120 and some programs won't run unless you say Y here. In particular, if
121 you want to run the DOS emulator dosemu under Linux (read the
122 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
123 you'll need to say Y here.
124
125 You can find documentation about IPC with "info ipc" and also in
126 section 6.4 of the Linux Programmer's Guide, available from
127 <http://www.tldp.org/guides.html>.
128
Eric W. Biedermana5494dc2007-02-14 00:34:06 -0800129config SYSVIPC_SYSCTL
130 bool
131 depends on SYSVIPC
132 depends on SYSCTL
133 default y
134
Linus Torvalds1da177e2005-04-16 15:20:36 -0700135config POSIX_MQUEUE
136 bool "POSIX Message Queues"
137 depends on NET && EXPERIMENTAL
138 ---help---
139 POSIX variant of message queues is a part of IPC. In POSIX message
140 queues every message has a priority which decides about succession
141 of receiving it by a process. If you want to compile and run
142 programs written e.g. for Solaris with use of its POSIX message
Robert P. J. Dayb0e37652007-05-09 07:25:13 +0200143 queues (functions mq_*) say Y here.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700144
145 POSIX message queues are visible as a filesystem called 'mqueue'
146 and can be mounted somewhere if you want to do filesystem
147 operations on message queues.
148
149 If unsure, say Y.
150
151config BSD_PROCESS_ACCT
152 bool "BSD Process Accounting"
153 help
154 If you say Y here, a user level program will be able to instruct the
155 kernel (via a special system call) to write process accounting
156 information to a file: whenever a process exits, information about
157 that process will be appended to the file by the kernel. The
158 information includes things such as creation time, owning user,
159 command name, memory usage, controlling terminal etc. (the complete
160 list is in the struct acct in <file:include/linux/acct.h>). It is
161 up to the user level program to do useful things with this
162 information. This is generally a good idea, so say Y.
163
164config BSD_PROCESS_ACCT_V3
165 bool "BSD Process Accounting version 3 file format"
166 depends on BSD_PROCESS_ACCT
167 default n
168 help
169 If you say Y here, the process accounting information is written
170 in a new file format that also logs the process IDs of each
171 process and it's parent. Note that this file format is incompatible
172 with previous v0/v1/v2 file formats, so you will need updated tools
173 for processing it. A preliminary version of these tools is available
S.Çağlar Onur37a4c942008-06-18 11:45:13 +0300174 at <http://www.gnu.org/software/acct/>.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700175
Shailabh Nagarc7572492006-07-14 00:24:40 -0700176config TASKSTATS
177 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
178 depends on NET
179 default n
180 help
181 Export selected statistics for tasks/processes through the
182 generic netlink interface. Unlike BSD process accounting, the
183 statistics are available during the lifetime of tasks/processes as
184 responses to commands. Like BSD accounting, they are sent to user
185 space on task exit.
186
187 Say N if unsure.
188
Shailabh Nagarca74e922006-07-14 00:24:36 -0700189config TASK_DELAY_ACCT
190 bool "Enable per-task delay accounting (EXPERIMENTAL)"
Shailabh Nagar6f449932006-07-14 00:24:41 -0700191 depends on TASKSTATS
Shailabh Nagarca74e922006-07-14 00:24:36 -0700192 help
193 Collect information on time spent by a task waiting for system
194 resources like cpu, synchronous block I/O completion and swapping
195 in pages. Such statistics can help in setting a task's priorities
196 relative to other tasks for cpu, io, rss limits etc.
197
198 Say N if unsure.
199
Alexey Dobriyan18f705f2007-02-10 01:46:44 -0800200config TASK_XACCT
201 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
202 depends on TASKSTATS
203 help
204 Collect extended task accounting data and send the data
205 to userland for processing over the taskstats interface.
206
207 Say N if unsure.
208
209config TASK_IO_ACCOUNTING
210 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
211 depends on TASK_XACCT
212 help
213 Collect information on the number of bytes of storage I/O which this
214 task has caused.
215
216 Say N if unsure.
217
Linus Torvalds1da177e2005-04-16 15:20:36 -0700218config AUDIT
219 bool "Auditing support"
Chris Wright804a6a492005-05-11 10:52:45 +0100220 depends on NET
Linus Torvalds1da177e2005-04-16 15:20:36 -0700221 help
222 Enable auditing infrastructure that can be used with another
223 kernel subsystem, such as SELinux (which requires this for
224 logging of avc messages output). Does not do system-call
225 auditing without CONFIG_AUDITSYSCALL.
226
227config AUDITSYSCALL
228 bool "Enable system-call auditing support"
Yuichi Nakamura1322b9d2007-11-10 19:21:34 +0900229 depends on AUDIT && (X86 || PPC || PPC64 || S390 || IA64 || UML || SPARC64|| SUPERH)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700230 default y if SECURITY_SELINUX
231 help
232 Enable low-overhead system-call auditing infrastructure that
233 can be used independently or with another kernel subsystem,
Amy Griffisf368c07d2006-04-07 16:55:56 -0400234 such as SELinux. To use audit's filesystem watch feature, please
235 ensure that INOTIFY is configured.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700236
Al Viro74c3cbe2007-07-22 08:04:18 -0400237config AUDIT_TREE
238 def_bool y
239 depends on AUDITSYSCALL && INOTIFY
240
Linus Torvalds1da177e2005-04-16 15:20:36 -0700241config IKCONFIG
Ross Birof2443ab2006-09-30 23:27:25 -0700242 tristate "Kernel .config support"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700243 ---help---
244 This option enables the complete Linux kernel ".config" file
245 contents to be saved in the kernel. It provides documentation
246 of which kernel options are used in a running kernel or in an
247 on-disk kernel. This information can be extracted from the kernel
248 image file with the script scripts/extract-ikconfig and used as
249 input to rebuild the current kernel or to build another kernel.
250 It can also be extracted from a running kernel by reading
251 /proc/config.gz if enabled (below).
252
253config IKCONFIG_PROC
254 bool "Enable access to .config through /proc/config.gz"
255 depends on IKCONFIG && PROC_FS
256 ---help---
257 This option enables access to the kernel configuration file
258 through /proc/config.gz.
259
Alistair John Strachan794543a2007-05-08 00:31:15 -0700260config LOG_BUF_SHIFT
261 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
262 range 12 21
Adrian Bunkf17a32e2008-04-29 00:58:58 -0700263 default 17
Alistair John Strachan794543a2007-05-08 00:31:15 -0700264 help
265 Select kernel log buffer size as a power of 2.
Adrian Bunkf17a32e2008-04-29 00:58:58 -0700266 Examples:
267 17 => 128 KB
268 16 => 64 KB
269 15 => 32 KB
270 14 => 16 KB
Alistair John Strachan794543a2007-05-08 00:31:15 -0700271 13 => 8 KB
272 12 => 4 KB
273
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800274#
275# Architectures with an unreliable sched_clock() should select this:
276#
277config HAVE_UNSTABLE_SCHED_CLOCK
278 bool
279
280config GROUP_SCHED
281 bool "Group CPU scheduler"
282 depends on EXPERIMENTAL
283 default n
284 help
285 This feature lets CPU scheduler recognize task groups and control CPU
286 bandwidth allocation to such task groups.
287 In order to create a group from arbitrary set of processes, use
288 CONFIG_CGROUPS. (See Control Group support.)
289
290config FAIR_GROUP_SCHED
291 bool "Group scheduling for SCHED_OTHER"
292 depends on GROUP_SCHED
293 default GROUP_SCHED
294
295config RT_GROUP_SCHED
296 bool "Group scheduling for SCHED_RR/FIFO"
297 depends on EXPERIMENTAL
298 depends on GROUP_SCHED
299 default n
300 help
301 This feature lets you explicitly allocate real CPU bandwidth
302 to users or control groups (depending on the "Basis for grouping tasks"
303 setting below. If enabled, it will also make it impossible to
304 schedule realtime tasks for non-root users until you allocate
305 realtime bandwidth for them.
306 See Documentation/scheduler/sched-rt-group.txt for more information.
307
308choice
309 depends on GROUP_SCHED
310 prompt "Basis for grouping tasks"
311 default USER_SCHED
312
313config USER_SCHED
314 bool "user id"
315 help
316 This option will choose userid as the basis for grouping
317 tasks, thus providing equal CPU bandwidth to each user.
318
319config CGROUP_SCHED
320 bool "Control groups"
321 depends on CGROUPS
322 help
323 This option allows you to create arbitrary task groups
324 using the "cgroup" pseudo filesystem and control
325 the cpu bandwidth allocated to each such task group.
326 Refer to Documentation/cgroups.txt for more information
327 on "cgroup" pseudo filesystem.
328
329endchoice
330
Li Zefan23964d22009-01-15 13:50:58 -0800331menuconfig CGROUPS
332 boolean "Control Group support"
Paul Menageddbcc7e2007-10-18 23:39:30 -0700333 help
Li Zefan23964d22009-01-15 13:50:58 -0800334 This option adds support for grouping sets of processes together, for
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800335 use with process control subsystems such as Cpusets, CFS, memory
336 controls or device isolation.
337 See
338 - Documentation/cpusets.txt (Cpusets)
339 - Documentation/scheduler/sched-design-CFS.txt (CFS)
340 - Documentation/cgroups/ (features for grouping, isolation)
341 - Documentation/controllers/ (features for resource control)
Paul Menageddbcc7e2007-10-18 23:39:30 -0700342
343 Say N if unsure.
344
Li Zefan23964d22009-01-15 13:50:58 -0800345if CGROUPS
346
Paul Menage006cb992007-10-18 23:39:43 -0700347config CGROUP_DEBUG
348 bool "Example debug cgroup subsystem"
349 depends on CGROUPS
Paul Menage418d7d82008-04-29 01:00:05 -0700350 default n
Paul Menage006cb992007-10-18 23:39:43 -0700351 help
352 This option enables a simple cgroup subsystem that
353 exports useful debugging information about the cgroups
Li Zefan23964d22009-01-15 13:50:58 -0800354 framework.
Paul Menage006cb992007-10-18 23:39:43 -0700355
Li Zefan23964d22009-01-15 13:50:58 -0800356 Say N if unsure.
Paul Menage006cb992007-10-18 23:39:43 -0700357
Serge E. Hallyn858d72e2007-10-18 23:39:45 -0700358config CGROUP_NS
Li Zefan23964d22009-01-15 13:50:58 -0800359 bool "Namespace cgroup subsystem"
360 depends on CGROUPS
361 help
362 Provides a simple namespace cgroup subsystem to
363 provide hierarchical naming of sets of namespaces,
364 for instance virtual servers and checkpoint/restart
365 jobs.
Serge E. Hallyn858d72e2007-10-18 23:39:45 -0700366
Matt Helsleydc52ddc2008-10-18 20:27:21 -0700367config CGROUP_FREEZER
Li Zefan23964d22009-01-15 13:50:58 -0800368 bool "Freezer cgroup subsystem"
369 depends on CGROUPS
370 help
371 Provides a way to freeze and unfreeze all tasks in a
Matt Helsleydc52ddc2008-10-18 20:27:21 -0700372 cgroup.
373
Serge E. Hallyn08ce5f12008-04-29 01:00:10 -0700374config CGROUP_DEVICE
375 bool "Device controller for cgroups"
376 depends on CGROUPS && EXPERIMENTAL
377 help
378 Provides a cgroup implementing whitelists for devices which
379 a process in the cgroup can mknod or open.
380
Linus Torvalds1da177e2005-04-16 15:20:36 -0700381config CPUSETS
382 bool "Cpuset support"
Paul Menage8793d852007-10-18 23:39:39 -0700383 depends on SMP && CGROUPS
Linus Torvalds1da177e2005-04-16 15:20:36 -0700384 help
Randy Dunlapd9fd8a62005-07-27 11:45:11 -0700385 This option will let you create and manage CPUSETs which
Linus Torvalds1da177e2005-04-16 15:20:36 -0700386 allow dynamically partitioning a system into sets of CPUs and
387 Memory Nodes and assigning tasks to run only within those sets.
388 This is primarily useful on large SMP or NUMA systems.
389
390 Say N if unsure.
391
Li Zefan23964d22009-01-15 13:50:58 -0800392config PROC_PID_CPUSET
393 bool "Include legacy /proc/<pid>/cpuset file"
394 depends on CPUSETS
395 default y
396
Srivatsa Vaddagirid842de82007-12-02 20:04:49 +0100397config CGROUP_CPUACCT
398 bool "Simple CPU accounting cgroup subsystem"
399 depends on CGROUPS
400 help
401 Provides a simple Resource Controller for monitoring the
Li Zefan23964d22009-01-15 13:50:58 -0800402 total CPU consumed by the tasks in a cgroup.
Srivatsa Vaddagirid842de82007-12-02 20:04:49 +0100403
Pavel Emelianove552b662008-02-07 00:13:49 -0800404config RESOURCE_COUNTERS
405 bool "Resource counters"
406 help
407 This option enables controller independent resource accounting
Li Zefan23964d22009-01-15 13:50:58 -0800408 infrastructure that works with cgroups.
Pavel Emelianove552b662008-02-07 00:13:49 -0800409 depends on CGROUPS
410
Balbir Singh00f0b822008-03-04 14:28:39 -0800411config CGROUP_MEM_RES_CTLR
412 bool "Memory Resource Controller for Control Groups"
413 depends on CGROUPS && RESOURCE_COUNTERS
Balbir Singhcf475ad2008-04-29 01:00:16 -0700414 select MM_OWNER
Balbir Singh00f0b822008-03-04 14:28:39 -0800415 help
KAMEZAWA Hiroyuki84ad6d72008-10-29 14:01:06 -0700416 Provides a memory resource controller that manages both anonymous
417 memory and page cache. (See Documentation/controllers/memory.txt)
Balbir Singh00f0b822008-03-04 14:28:39 -0800418
419 Note that setting this option increases fixed memory overhead
KAMEZAWA Hiroyuki84ad6d72008-10-29 14:01:06 -0700420 associated with each page of memory in the system. By this,
421 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
422 usage tracking struct at boot. Total amount of this is printed out
423 at boot.
Balbir Singh00f0b822008-03-04 14:28:39 -0800424
425 Only enable when you're ok with these trade offs and really
KAMEZAWA Hiroyuki84ad6d72008-10-29 14:01:06 -0700426 sure you need the memory resource controller. Even when you enable
427 this, you can set "cgroup_disable=memory" at your boot option to
428 disable memory resource controller and you can avoid overheads.
Li Zefanc9d54092009-01-07 18:07:35 -0800429 (and lose benefits of memory resource controller)
Balbir Singh00f0b822008-03-04 14:28:39 -0800430
Balbir Singhcf475ad2008-04-29 01:00:16 -0700431 This config option also selects MM_OWNER config option, which
432 could in turn add some fork/exit overhead.
433
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -0800434config CGROUP_MEM_RES_CTLR_SWAP
435 bool "Memory Resource Controller Swap Extension(EXPERIMENTAL)"
436 depends on CGROUP_MEM_RES_CTLR && SWAP && EXPERIMENTAL
437 help
438 Add swap management feature to memory resource controller. When you
439 enable this, you can limit mem+swap usage per cgroup. In other words,
440 when you disable this, memory resource controller has no cares to
441 usage of swap...a process can exhaust all of the swap. This extension
442 is useful when you want to avoid exhaustion swap but this itself
443 adds more overheads and consumes memory for remembering information.
444 Especially if you use 32bit system or small memory system, please
445 be careful about enabling this. When memory resource controller
446 is disabled by boot option, this will be automatically disabled and
447 there will be no overhead from this. Even when you set this config=y,
448 if boot option "noswapaccount" is set, swap will not be accounted.
449
Li Zefan23964d22009-01-15 13:50:58 -0800450endif # CGROUPS
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -0800451
Li Zefan23964d22009-01-15 13:50:58 -0800452config MM_OWNER
453 bool
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800454
Kay Sievers88a22c92006-09-14 11:23:28 +0200455config SYSFS_DEPRECATED
Ingo Molnard47846c2008-03-04 14:54:47 +0100456 bool
457
458config SYSFS_DEPRECATED_V2
Kay Sieversfce3e802008-11-01 14:03:00 +0100459 bool "Create deprecated sysfs layout for older userspace tools"
Randy Dunlap9148fe82007-12-31 10:05:34 -0800460 depends on SYSFS
Kay Sievers88a22c92006-09-14 11:23:28 +0200461 default y
Ingo Molnard47846c2008-03-04 14:54:47 +0100462 select SYSFS_DEPRECATED
Kay Sievers88a22c92006-09-14 11:23:28 +0200463 help
Kay Sieversfce3e802008-11-01 14:03:00 +0100464 This option switches the layout of sysfs to the deprecated
465 version.
Kay Sievers88a22c92006-09-14 11:23:28 +0200466
Kay Sieversfce3e802008-11-01 14:03:00 +0100467 The current sysfs layout features a unified device tree at
468 /sys/devices/, which is able to express a hierarchy between
469 class devices. If the deprecated option is set to Y, the
470 unified device tree is split into a bus device tree at
471 /sys/devices/ and several individual class device trees at
472 /sys/class/. The class and bus devices will be connected by
473 "<subsystem>:<name>" and the "device" links. The "block"
474 class devices, will not show up in /sys/class/block/. Some
475 subsystems will suppress the creation of some devices which
476 depend on the unified device tree.
Kay Sievers88a22c92006-09-14 11:23:28 +0200477
Kay Sieversfce3e802008-11-01 14:03:00 +0100478 This option is not a pure compatibility option that can
479 be safely enabled on newer distributions. It will change the
480 layout of sysfs to the non-extensible deprecated version,
481 and disable some features, which can not be exported without
482 confusing older userspace tools. Since 2007/2008 all major
483 distributions do not enable this option, and ship no tools which
484 depend on the deprecated layout or this option.
485
486 If you are using a new kernel on an older distribution, or use
487 older userspace tools, you might need to say Y here. Do not say Y,
488 if the original kernel, that came with your distribution, has
489 this option set to N.
Kay Sievers88a22c92006-09-14 11:23:28 +0200490
Jens Axboeb86ff9812006-03-23 19:56:55 +0100491config RELAY
492 bool "Kernel->user space relay support (formerly relayfs)"
493 help
494 This option enables support for relay interface support in
495 certain file systems (such as debugfs).
496 It is designed to provide an efficient mechanism for tools and
497 facilities to relay large amounts of data from kernel space to
498 user space.
499
500 If unsure, say N.
501
Pavel Emelyanovc5289a692008-02-08 04:18:19 -0800502config NAMESPACES
503 bool "Namespaces support" if EMBEDDED
504 default !EMBEDDED
505 help
506 Provides the way to make tasks work with different objects using
507 the same id. For example same IPC id may refer to different objects
508 or same user id or pid may refer to different tasks when used in
509 different namespaces.
510
Pavel Emelyanov58bfdd6d2008-02-08 04:18:21 -0800511config UTS_NS
512 bool "UTS namespace"
513 depends on NAMESPACES
514 help
515 In this namespace tasks see different info provided with the
516 uname() system call
517
Pavel Emelyanovae5e1b22008-02-08 04:18:22 -0800518config IPC_NS
519 bool "IPC namespace"
520 depends on NAMESPACES && SYSVIPC
521 help
522 In this namespace tasks work with IPC ids which correspond to
523 different IPC objects in different namespaces
524
Pavel Emelyanovaee16ce2008-02-08 04:18:23 -0800525config USER_NS
526 bool "User namespace (EXPERIMENTAL)"
527 depends on NAMESPACES && EXPERIMENTAL
528 help
529 This allows containers, i.e. vservers, to use user namespaces
530 to provide different user info for different servers.
531 If unsure, say N.
532
Pavel Emelyanov74bd59b2008-02-08 04:18:24 -0800533config PID_NS
534 bool "PID Namespaces (EXPERIMENTAL)"
535 default n
536 depends on NAMESPACES && EXPERIMENTAL
537 help
Heikki Orsila12d2b8f2008-07-06 15:48:02 +0300538 Support process id namespaces. This allows having multiple
Pavel Emelyanov74bd59b2008-02-08 04:18:24 -0800539 process with the same pid as long as they are in different
540 pid namespaces. This is a building block of containers.
541
542 Unless you want to work with an experimental feature
543 say N here.
544
Dimitri Gorokhovikf9916332007-03-06 01:42:17 -0800545config BLK_DEV_INITRD
546 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
547 depends on BROKEN || !FRV
548 help
549 The initial RAM filesystem is a ramfs which is loaded by the
550 boot loader (loadlin or lilo) and that is mounted as root
551 before the normal boot procedure. It is typically used to
552 load modules needed to mount the "real" root file system,
553 etc. See <file:Documentation/initrd.txt> for details.
554
555 If RAM disk support (BLK_DEV_RAM) is also included, this
556 also enables initial RAM disk (initrd) support and adds
557 15 Kbytes (more on some other architectures) to the kernel size.
558
559 If unsure say Y.
560
Jean-Paul Samanc33df4e2007-02-10 01:44:43 -0800561if BLK_DEV_INITRD
562
Sam Ravnborgdbec4862005-08-10 20:44:50 +0200563source "usr/Kconfig"
564
Jean-Paul Samanc33df4e2007-02-10 01:44:43 -0800565endif
566
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800567config CC_OPTIMIZE_FOR_SIZE
Ingo Molnar96fffeb2008-04-28 01:39:43 +0200568 bool "Optimize for size"
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800569 default y
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800570 help
571 Enabling this option will pass "-Os" instead of "-O2" to gcc
572 resulting in a smaller kernel.
573
jkacur775a7222008-07-16 00:31:16 +0200574 If unsure, say Y.
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800575
Randy Dunlap08470622006-09-30 23:28:13 -0700576config SYSCTL
577 bool
578
Linus Torvalds1da177e2005-04-16 15:20:36 -0700579menuconfig EMBEDDED
580 bool "Configure standard kernel features (for small systems)"
581 help
582 This option allows certain base kernel options and settings
583 to be disabled or tweaked. This is for specialized
584 environments which can tolerate a "non-standard" kernel.
585 Only use this if you really know what you are doing.
586
Chuck Ebbertae81f9e2006-09-16 12:15:53 -0700587config UID16
588 bool "Enable 16-bit UID system calls" if EMBEDDED
David S. Miller09337f52008-04-26 03:17:12 -0700589 depends on ARM || BLACKFIN || CRIS || FRV || H8300 || X86_32 || M68K || (S390 && !64BIT) || SUPERH || SPARC32 || (SPARC64 && COMPAT) || UML || (X86_64 && IA32_EMULATION)
Chuck Ebbertae81f9e2006-09-16 12:15:53 -0700590 default y
591 help
592 This enables the legacy 16-bit UID syscall wrappers.
593
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700594config SYSCTL_SYSCALL
Randy Dunlap08470622006-09-30 23:28:13 -0700595 bool "Sysctl syscall support" if EMBEDDED
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800596 default y
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700597 select SYSCTL
598 ---help---
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800599 sys_sysctl uses binary paths that have been found challenging
600 to properly maintain and use. The interface in /proc/sys
601 using paths with ascii names is now the primary path to this
602 information.
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700603
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800604 Almost nothing using the binary sysctl interface so if you are
605 trying to save some space it is probably safe to disable this,
606 making your kernel marginally smaller.
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700607
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800608 If unsure say Y here.
Chuck Ebbertae81f9e2006-09-16 12:15:53 -0700609
Linus Torvalds1da177e2005-04-16 15:20:36 -0700610config KALLSYMS
Jesper Juhl979c6a12006-12-12 19:25:11 +0100611 bool "Load all symbols for debugging/ksymoops" if EMBEDDED
Linus Torvalds1da177e2005-04-16 15:20:36 -0700612 default y
613 help
614 Say Y here to let the kernel print out symbolic crash information and
615 symbolic stack backtraces. This increases the size of the kernel
616 somewhat, as all symbols have to be loaded into the kernel image.
617
618config KALLSYMS_ALL
619 bool "Include all symbols in kallsyms"
620 depends on DEBUG_KERNEL && KALLSYMS
621 help
622 Normally kallsyms only contains the symbols of functions, for nicer
623 OOPS messages. Some debuggers can use kallsyms for other
Jesper Juhlf9f97bc2005-07-20 05:43:05 +0200624 symbols too: say Y here to include all symbols, if you need them
625 and you don't care about adding 300k to the size of your kernel.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700626
627 Say N.
628
Jan Beulich9bb48242008-12-16 11:30:08 +0000629config KALLSYMS_STRIP_GENERATED
630 bool "Strip machine generated symbols from kallsyms"
631 depends on KALLSYMS_ALL
632 default y
633 help
634 Say N if you want kallsyms to retain even machine generated symbols.
635
Linus Torvalds1da177e2005-04-16 15:20:36 -0700636config KALLSYMS_EXTRA_PASS
637 bool "Do an extra kallsyms pass"
638 depends on KALLSYMS
639 help
640 If kallsyms is not working correctly, the build will fail with
641 inconsistent kallsyms data. If that occurs, log a bug report and
642 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
643 Always say N here unless you find a bug in kallsyms, which must be
644 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
645 you wait for kallsyms to be fixed.
646
Matt Mackalld59745c2005-05-01 08:59:02 -0700647
Greg Kroah-Hartman712f47c2005-11-16 11:27:07 -0800648config HOTPLUG
649 bool "Support for hot-pluggable devices" if EMBEDDED
650 default y
651 help
652 This option is provided for the case where no hotplug or uevent
653 capabilities is wanted by the kernel. You should only consider
654 disabling this option for embedded systems that do not use modules, a
655 dynamic /dev tree, or dynamic device discovery. Just say Y.
656
Matt Mackalld59745c2005-05-01 08:59:02 -0700657config PRINTK
658 default y
659 bool "Enable support for printk" if EMBEDDED
660 help
661 This option enables normal printk support. Removing it
662 eliminates most of the message strings from the kernel image
663 and makes the kernel more or less silent. As this makes it
664 very difficult to diagnose system problems, saying N here is
665 strongly discouraged.
666
Matt Mackallc8538a72005-05-01 08:59:01 -0700667config BUG
668 bool "BUG() support" if EMBEDDED
669 default y
670 help
671 Disabling this option eliminates support for BUG and WARN, reducing
672 the size of your kernel image and potentially quietly ignoring
673 numerous fatal conditions. You should only consider disabling this
674 option for embedded systems with no facilities for reporting errors.
675 Just say Y.
676
Matt Mackall708e9a72006-01-08 01:05:25 -0800677config ELF_CORE
678 default y
679 bool "Enable ELF core dumps" if EMBEDDED
680 help
681 Enable support for generating core dumps. Disabling saves about 4k.
682
Stas Sergeeve5e1d3c2008-05-07 12:39:56 +0200683config PCSPKR_PLATFORM
684 bool "Enable PC-Speaker support" if EMBEDDED
685 depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
686 default y
687 help
688 This option allows to disable the internal PC-Speaker
689 support, saving some memory.
690
Ingo Molnar32a93232008-02-06 22:39:44 +0100691config COMPAT_BRK
692 bool "Disable heap randomization"
693 default y
694 help
695 Randomizing heap placement makes heap exploits harder, but it
696 also breaks ancient binaries (including anything libc5 based).
697 This option changes the bootup default to heap randomization
698 disabled, and can be overriden runtime by setting
699 /proc/sys/kernel/randomize_va_space to 2.
700
Ingo Molnar166124f2008-02-09 23:24:09 +0100701 On non-ancient distros (post-2000 ones) N is usually a safe choice.
Ingo Molnar32a93232008-02-06 22:39:44 +0100702
Linus Torvalds1da177e2005-04-16 15:20:36 -0700703config BASE_FULL
704 default y
705 bool "Enable full-sized data structures for core" if EMBEDDED
706 help
707 Disabling this option reduces the size of miscellaneous core
708 kernel data structures. This saves memory on small machines,
709 but may reduce performance.
710
711config FUTEX
712 bool "Enable futex support" if EMBEDDED
713 default y
Ingo Molnar23f78d4a2006-06-27 02:54:53 -0700714 select RT_MUTEXES
Linus Torvalds1da177e2005-04-16 15:20:36 -0700715 help
716 Disabling this option will cause the kernel to be built without
717 support for "fast userspace mutexes". The resulting kernel may not
718 run glibc-based applications correctly.
719
Davide Libenzi5dc8bf82007-05-10 22:23:11 -0700720config ANON_INODES
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700721 bool
Davide Libenzi5dc8bf82007-05-10 22:23:11 -0700722
Linus Torvalds1da177e2005-04-16 15:20:36 -0700723config EPOLL
724 bool "Enable eventpoll support" if EMBEDDED
725 default y
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700726 select ANON_INODES
Linus Torvalds1da177e2005-04-16 15:20:36 -0700727 help
728 Disabling this option will cause the kernel to be built without
729 support for epoll family of system calls.
730
Davide Libenzifba2afa2007-05-10 22:23:13 -0700731config SIGNALFD
732 bool "Enable signalfd() system call" if EMBEDDED
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700733 select ANON_INODES
Davide Libenzifba2afa2007-05-10 22:23:13 -0700734 default y
735 help
736 Enable the signalfd() system call that allows to receive signals
737 on a file descriptor.
738
739 If unsure, say Y.
740
Davide Libenzib215e282007-05-10 22:23:16 -0700741config TIMERFD
742 bool "Enable timerfd() system call" if EMBEDDED
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700743 select ANON_INODES
Davide Libenzib215e282007-05-10 22:23:16 -0700744 default y
745 help
746 Enable the timerfd() system call that allows to receive timer
747 events on a file descriptor.
748
749 If unsure, say Y.
750
Davide Libenzie1ad7462007-05-10 22:23:19 -0700751config EVENTFD
752 bool "Enable eventfd() system call" if EMBEDDED
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700753 select ANON_INODES
Davide Libenzie1ad7462007-05-10 22:23:19 -0700754 default y
755 help
756 Enable the eventfd() system call that allows to receive both
757 kernel notification (ie. KAIO) or userspace notifications.
758
759 If unsure, say Y.
760
Linus Torvalds1da177e2005-04-16 15:20:36 -0700761config SHMEM
762 bool "Use full shmem filesystem" if EMBEDDED
763 default y
764 depends on MMU
765 help
766 The shmem is an internal filesystem used to manage shared memory.
767 It is backed by swap and manages resource limits. It is also exported
768 to userspace as tmpfs if TMPFS is enabled. Disabling this
769 option replaces shmem and tmpfs with the much simpler ramfs code,
770 which may be appropriate on small systems without swap.
771
Thomas Petazzoniebf3f092008-10-15 22:05:12 -0700772config AIO
773 bool "Enable AIO support" if EMBEDDED
774 default y
775 help
776 This option enables POSIX asynchronous I/O which may by used
777 by some high performance threaded applications. Disabling
778 this option saves about 7k.
779
Christoph Lameterf8891e52006-06-30 01:55:45 -0700780config VM_EVENT_COUNTERS
781 default y
782 bool "Enable VM event counters for /proc/vmstat" if EMBEDDED
783 help
Paul Jackson2aea4fb2006-12-22 01:06:10 -0800784 VM event counters are needed for event counts to be shown.
785 This option allows the disabling of the VM event counters
786 on EMBEDDED systems. /proc/vmstat will only show page counts
787 if VM event counters are disabled.
Christoph Lameterf8891e52006-06-30 01:55:45 -0700788
Thomas Petazzoni3d137312008-08-19 10:28:24 +0200789config PCI_QUIRKS
790 default y
Geert Uytterhoeven61cfc7e2008-10-22 08:53:25 +0200791 bool "Enable PCI quirk workarounds" if EMBEDDED
792 depends on PCI
Thomas Petazzoni3d137312008-08-19 10:28:24 +0200793 help
794 This enables workarounds for various PCI chipset
795 bugs/quirks. Disable this only if your target machine is
796 unaffected by PCI quirks.
797
Christoph Lameter41ecc552007-05-09 02:32:44 -0700798config SLUB_DEBUG
799 default y
800 bool "Enable SLUB debugging support" if EMBEDDED
Christoph Lameterf6acb632008-04-29 16:16:06 -0700801 depends on SLUB && SYSFS
Christoph Lameter41ecc552007-05-09 02:32:44 -0700802 help
803 SLUB has extensive debug support features. Disabling these can
804 result in significant savings in code size. This also disables
805 SLUB sysfs support. /sys/slab will not exist and there will be
806 no support for cache validation etc.
807
Christoph Lameter81819f02007-05-06 14:49:36 -0700808choice
809 prompt "Choose SLAB allocator"
Christoph Lametera0acd822007-07-17 04:03:32 -0700810 default SLUB
Christoph Lameter81819f02007-05-06 14:49:36 -0700811 help
812 This option allows to select a slab allocator.
813
814config SLAB
815 bool "SLAB"
816 help
817 The regular slab allocator that is established and known to work
Christoph Lameter34013882007-05-09 02:32:47 -0700818 well in all environments. It organizes cache hot objects in
Simon Arlott02f56212008-11-05 22:18:19 +0000819 per cpu and per node queues.
Christoph Lameter81819f02007-05-06 14:49:36 -0700820
821config SLUB
Christoph Lameter81819f02007-05-06 14:49:36 -0700822 bool "SLUB (Unqueued Allocator)"
823 help
824 SLUB is a slab allocator that minimizes cache line usage
825 instead of managing queues of cached objects (SLAB approach).
826 Per cpu caching is realized using slabs of objects instead
827 of queues of objects. SLUB can use memory efficiently
Simon Arlott02f56212008-11-05 22:18:19 +0000828 and has enhanced diagnostics. SLUB is the default choice for
829 a slab allocator.
Christoph Lameter81819f02007-05-06 14:49:36 -0700830
831config SLOB
Paul Mundt84a01c22007-07-15 23:38:24 -0700832 depends on EMBEDDED
Christoph Lameter81819f02007-05-06 14:49:36 -0700833 bool "SLOB (Simple Allocator)"
834 help
Matt Mackall37291452008-02-04 22:29:38 -0800835 SLOB replaces the stock allocator with a drastically simpler
836 allocator. SLOB is generally more space efficient but
837 does not perform as well on large systems.
Christoph Lameter81819f02007-05-06 14:49:36 -0700838
839endchoice
840
Mathieu Desnoyers125e5642008-02-02 15:10:36 -0500841config PROFILING
842 bool "Profiling support (EXPERIMENTAL)"
843 help
844 Say Y here to enable the extended profiling support mechanisms used
845 by profilers such as OProfile.
846
Ingo Molnar5f87f112008-07-23 14:15:22 +0200847#
848# Place an empty function call at each tracepoint site. Can be
849# dynamically changed for a probe function.
850#
Mathieu Desnoyers97e1c182008-07-18 12:16:16 -0400851config TRACEPOINTS
Ingo Molnar5f87f112008-07-23 14:15:22 +0200852 bool
Mathieu Desnoyers97e1c182008-07-18 12:16:16 -0400853
Mathieu Desnoyers125e5642008-02-02 15:10:36 -0500854config MARKERS
855 bool "Activate markers"
Mathieu Desnoyersc1df1bd2008-11-14 17:47:39 -0500856 depends on TRACEPOINTS
Mathieu Desnoyers125e5642008-02-02 15:10:36 -0500857 help
858 Place an empty function call at each marker site. Can be
859 dynamically changed for a probe function.
860
Mathieu Desnoyersfb32e032008-02-02 15:10:33 -0500861source "arch/Kconfig"
862
Linus Torvalds1da177e2005-04-16 15:20:36 -0700863endmenu # General setup
864
Dmitry Baryshkovee7e5512008-06-29 14:18:46 +0400865config HAVE_GENERIC_DMA_COHERENT
866 bool
867 default n
868
Linus Torvalds158a9622008-01-02 13:04:48 -0800869config SLABINFO
870 bool
871 depends on PROC_FS
Christoph Lameter0f389ec2008-04-14 18:53:02 +0300872 depends on SLAB || SLUB_DEBUG
Linus Torvalds158a9622008-01-02 13:04:48 -0800873 default y
874
Chuck Ebbertae81f9e2006-09-16 12:15:53 -0700875config RT_MUTEXES
876 boolean
877 select PLIST
878
Linus Torvalds1da177e2005-04-16 15:20:36 -0700879config BASE_SMALL
880 int
881 default 0 if BASE_FULL
882 default 1 if !BASE_FULL
883
Jan Engelhardt66da5732007-07-15 23:39:29 -0700884menuconfig MODULES
Linus Torvalds1da177e2005-04-16 15:20:36 -0700885 bool "Enable loadable module support"
886 help
887 Kernel modules are small pieces of compiled code which can
888 be inserted in the running kernel, rather than being
889 permanently built into the kernel. You use the "modprobe"
890 tool to add (and sometimes remove) them. If you say Y here,
891 many parts of the kernel can be built as modules (by
892 answering M instead of Y where indicated): this is most
893 useful for infrequently used options which are not required
894 for booting. For more information, see the man pages for
895 modprobe, lsmod, modinfo, insmod and rmmod.
896
897 If you say Y here, you will need to run "make
898 modules_install" to put the modules under /lib/modules/
899 where modprobe can find them (you may need to be root to do
900 this).
901
902 If unsure, say Y.
903
Robert P. J. Day0b0de142008-08-04 13:31:32 -0400904if MODULES
905
Linus Torvalds826e4502008-05-04 17:04:16 -0700906config MODULE_FORCE_LOAD
907 bool "Forced module loading"
Linus Torvalds826e4502008-05-04 17:04:16 -0700908 default n
909 help
Rusty Russell91e37a72008-05-09 16:25:28 +1000910 Allow loading of modules without version information (ie. modprobe
911 --force). Forced module loading sets the 'F' (forced) taint flag and
912 is usually a really bad idea.
Linus Torvalds826e4502008-05-04 17:04:16 -0700913
Linus Torvalds1da177e2005-04-16 15:20:36 -0700914config MODULE_UNLOAD
915 bool "Module unloading"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700916 help
917 Without this option you will not be able to unload any
918 modules (note that some modules may not be unloadable
Denys Vlasenkof7f5b672008-07-22 19:24:26 -0500919 anyway), which makes your kernel smaller, faster
920 and simpler. If unsure, say Y.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700921
922config MODULE_FORCE_UNLOAD
923 bool "Forced module unloading"
924 depends on MODULE_UNLOAD && EXPERIMENTAL
925 help
926 This option allows you to force a module to unload, even if the
927 kernel believes it is unsafe: the kernel will remove the module
928 without waiting for anyone to stop using it (using the -f option to
929 rmmod). This is mainly for kernel developers and desperate users.
930 If unsure, say N.
931
Linus Torvalds1da177e2005-04-16 15:20:36 -0700932config MODVERSIONS
Sam Ravnborg0d541642005-12-26 23:04:02 +0100933 bool "Module versioning support"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700934 help
935 Usually, you have to use modules compiled with your kernel.
936 Saying Y here makes it sometimes possible to use modules
937 compiled for different kernels, by adding enough information
938 to the modules to (hopefully) spot any changes which would
939 make them incompatible with the kernel you are running. If
940 unsure, say N.
941
942config MODULE_SRCVERSION_ALL
943 bool "Source checksum for all modules"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700944 help
945 Modules which contain a MODULE_VERSION get an extra "srcversion"
946 field inserted into their modinfo section, which contains a
947 sum of the source files which made it. This helps maintainers
948 see exactly which source was used to build a module (since
949 others sometimes change the module source without updating
950 the version). With this option, such a "srcversion" field
951 will be created for all modules. If unsure, say N.
952
Robert P. J. Day0b0de142008-08-04 13:31:32 -0400953endif # MODULES
954
Rusty Russell98a79d62008-12-13 21:19:41 +1030955config INIT_ALL_POSSIBLE
956 bool
957 help
958 Back when each arch used to define their own cpu_online_map and
959 cpu_possible_map, some of them chose to initialize cpu_possible_map
960 with all 1s, and others with all 0s. When they were centralised,
961 it was better to provide this option than to break all the archs
962 and have several arch maintainers persuing me down dark alleys.
963
Linus Torvalds1da177e2005-04-16 15:20:36 -0700964config STOP_MACHINE
965 bool
966 default y
967 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
968 help
969 Need stop_machine() primitive.
Jens Axboe3a65dfe2005-11-04 08:43:35 +0100970
Jens Axboe3a65dfe2005-11-04 08:43:35 +0100971source "block/Kconfig"
Avi Kivitye98c3202007-10-16 23:27:31 -0700972
973config PREEMPT_NOTIFIERS
974 bool
Paul E. McKenneye260be62008-01-25 21:08:24 +0100975
Ingo Molnar12d79ba2008-12-25 09:31:28 +0100976choice
977 prompt "RCU Implementation"
978 default CLASSIC_RCU
979
Paul E. McKenneye260be62008-01-25 21:08:24 +0100980config CLASSIC_RCU
Ingo Molnar12d79ba2008-12-25 09:31:28 +0100981 bool "Classic RCU"
Paul E. McKenneye260be62008-01-25 21:08:24 +0100982 help
983 This option selects the classic RCU implementation that is
984 designed for best read-side performance on non-realtime
Ingo Molnar12d79ba2008-12-25 09:31:28 +0100985 systems.
986
987 Select this option if you are unsure.
988
989config TREE_RCU
990 bool "Tree-based hierarchical RCU"
991 help
992 This option selects the RCU implementation that is
993 designed for very large SMP system with hundreds or
994 thousands of CPUs.
995
996config PREEMPT_RCU
997 bool "Preemptible RCU"
998 depends on PREEMPT
999 help
1000 This option reduces the latency of the kernel by making certain
1001 RCU sections preemptible. Normally RCU code is non-preemptible, if
1002 this option is selected then read-only RCU sections become
1003 preemptible. This helps latency, but may expose bugs due to
1004 now-naive assumptions about each RCU read-side critical section
1005 remaining on a given CPU through its execution.
1006
1007endchoice
1008
1009config RCU_TRACE
1010 bool "Enable tracing for RCU"
1011 depends on TREE_RCU || PREEMPT_RCU
1012 help
1013 This option provides tracing in RCU which presents stats
1014 in debugfs for debugging RCU implementation.
1015
1016 Say Y here if you want to enable RCU tracing
1017 Say N if you are unsure.
1018
1019config RCU_FANOUT
1020 int "Tree-based hierarchical RCU fanout value"
1021 range 2 64 if 64BIT
1022 range 2 32 if !64BIT
1023 depends on TREE_RCU
1024 default 64 if 64BIT
1025 default 32 if !64BIT
1026 help
1027 This option controls the fanout of hierarchical implementations
1028 of RCU, allowing RCU to work efficiently on machines with
1029 large numbers of CPUs. This value must be at least the cube
1030 root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
1031 systems and up to 262,144 for 64-bit systems.
1032
1033 Select a specific number if testing RCU itself.
1034 Take the default if unsure.
1035
1036config RCU_FANOUT_EXACT
1037 bool "Disable tree-based hierarchical RCU auto-balancing"
1038 depends on TREE_RCU
1039 default n
1040 help
1041 This option forces use of the exact RCU_FANOUT value specified,
1042 regardless of imbalances in the hierarchy. This is useful for
1043 testing RCU itself, and might one day be useful on systems with
1044 strong NUMA behavior.
1045
1046 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
1047
1048 Say N if unsure.
1049
Paul E. McKenney64db4cf2008-12-18 21:55:32 +01001050config TREE_RCU_TRACE
1051 def_bool RCU_TRACE && TREE_RCU
1052 select DEBUG_FS
Paul E. McKenneye260be62008-01-25 21:08:24 +01001053 help
Paul E. McKenney64db4cf2008-12-18 21:55:32 +01001054 This option provides tracing for the TREE_RCU implementation,
1055 permitting Makefile to trivially select kernel/rcutree_trace.c.
1056
1057config PREEMPT_RCU_TRACE
1058 def_bool RCU_TRACE && PREEMPT_RCU
1059 select DEBUG_FS
1060 help
1061 This option provides tracing for the PREEMPT_RCU implementation,
1062 permitting Makefile to trivially select kernel/rcupreempt_trace.c.