blob: 42bf914b325af249784468022b70db98f54f0b7a [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
Peter Oberparleiterb99b87f72009-06-17 16:28:03 -070019config CONSTRUCTORS
20 bool
21 depends on !UML
22 default y
23
Al Boldiff0cfc62007-07-31 00:39:23 -070024menu "General setup"
Linus Torvalds1da177e2005-04-16 15:20:36 -070025
26config EXPERIMENTAL
27 bool "Prompt for development and/or incomplete code/drivers"
28 ---help---
29 Some of the various things that Linux supports (such as network
30 drivers, file systems, network protocols, etc.) can be in a state
31 of development where the functionality, stability, or the level of
32 testing is not yet high enough for general use. This is usually
33 known as the "alpha-test" phase among developers. If a feature is
34 currently in alpha-test, then the developers usually discourage
35 uninformed widespread use of this feature by the general public to
36 avoid "Why doesn't this work?" type mail messages. However, active
37 testing and use of these systems is welcomed. Just be aware that it
38 may not meet the normal level of reliability or it may fail to work
39 in some special cases. Detailed bug reports from people familiar
40 with the kernel internals are usually welcomed by the developers
41 (before submitting bug reports, please read the documents
42 <file:README>, <file:MAINTAINERS>, <file:REPORTING-BUGS>,
43 <file:Documentation/BUG-HUNTING>, and
44 <file:Documentation/oops-tracing.txt> in the kernel source).
45
46 This option will also make obsoleted drivers available. These are
47 drivers that have been replaced by something else, and/or are
48 scheduled to be removed in a future kernel release.
49
50 Unless you intend to help test and develop a feature or driver that
51 falls into this category, or you have a situation that requires
52 using these features, you should probably say N here, which will
53 cause the configurator to present you with fewer choices. If
54 you say Y here, you will be offered the choice of using features or
55 drivers that are currently considered to be in the alpha-test phase.
56
Linus Torvalds1da177e2005-04-16 15:20:36 -070057config BROKEN
58 bool
Linus Torvalds1da177e2005-04-16 15:20:36 -070059
60config BROKEN_ON_SMP
61 bool
62 depends on BROKEN || !SMP
63 default y
64
65config LOCK_KERNEL
66 bool
67 depends on SMP || PREEMPT
68 default y
69
70config INIT_ENV_ARG_LIMIT
71 int
Adrian Bunkdd673bc2006-06-30 01:55:51 -070072 default 32 if !UML
73 default 128 if UML
Linus Torvalds1da177e2005-04-16 15:20:36 -070074 help
Randy Dunlap34ad92c22005-10-30 15:01:46 -080075 Maximum of each of the number of arguments and environment
76 variables passed to init from the kernel command line.
Linus Torvalds1da177e2005-04-16 15:20:36 -070077
Linus Torvalds1da177e2005-04-16 15:20:36 -070078
79config LOCALVERSION
80 string "Local version - append to kernel release"
81 help
82 Append an extra string to the end of your kernel version.
83 This will show up when you type uname, for example.
84 The string you set here will be appended after the contents of
85 any files with a filename matching localversion* in your
86 object and source tree, in that order. Your total string can
87 be a maximum of 64 characters.
88
Ryan Andersonaaebf432005-07-31 04:57:49 -040089config LOCALVERSION_AUTO
90 bool "Automatically append version information to the version string"
91 default y
92 help
93 This will try to automatically determine if the current tree is a
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020094 release tree by looking for git tags that belong to the current
95 top of tree revision.
Ryan Andersonaaebf432005-07-31 04:57:49 -040096
97 A string of the format -gxxxxxxxx will be added to the localversion
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020098 if a git-based tree is found. The string generated by this will be
Ryan Andersonaaebf432005-07-31 04:57:49 -040099 appended after any matching localversion* files, and after the value
Robert P. J. Day6e5a5422007-05-01 23:08:11 +0200100 set in CONFIG_LOCALVERSION.
Ryan Andersonaaebf432005-07-31 04:57:49 -0400101
Robert P. J. Day6e5a5422007-05-01 23:08:11 +0200102 (The actual string used here is the first eight characters produced
103 by running the command:
104
105 $ git rev-parse --verify HEAD
106
107 which is done within the script "scripts/setlocalversion".)
Ryan Andersonaaebf432005-07-31 04:57:49 -0400108
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800109config HAVE_KERNEL_GZIP
110 bool
111
112config HAVE_KERNEL_BZIP2
113 bool
114
115config HAVE_KERNEL_LZMA
116 bool
117
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800118config HAVE_KERNEL_LZO
119 bool
120
Alain Knaff30d65db2009-01-04 22:46:17 +0100121choice
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800122 prompt "Kernel compression mode"
123 default KERNEL_GZIP
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800124 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_LZO
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800125 help
Alain Knaff30d65db2009-01-04 22:46:17 +0100126 The linux kernel is a kind of self-extracting executable.
127 Several compression algorithms are available, which differ
128 in efficiency, compression and decompression speed.
129 Compression speed is only relevant when building a kernel.
130 Decompression speed is relevant at each boot.
131
132 If you have any problems with bzip2 or lzma compressed
133 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
134 version of this functionality (bzip2 only), for 2.4, was
135 supplied by Christian Ludwig)
136
137 High compression options are mostly useful for users, who
138 are low on disk space (embedded systems), but for whom ram
139 size matters less.
140
141 If in doubt, select 'gzip'
142
143config KERNEL_GZIP
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800144 bool "Gzip"
145 depends on HAVE_KERNEL_GZIP
146 help
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800147 The old and tried gzip compression. It provides a good balance
148 between compression ratio and decompression speed.
Alain Knaff30d65db2009-01-04 22:46:17 +0100149
150config KERNEL_BZIP2
151 bool "Bzip2"
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800152 depends on HAVE_KERNEL_BZIP2
Alain Knaff30d65db2009-01-04 22:46:17 +0100153 help
154 Its compression ratio and speed is intermediate.
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800155 Decompression speed is slowest among the three. The kernel
156 size is about 10% smaller with bzip2, in comparison to gzip.
157 Bzip2 uses a large amount of memory. For modern kernels you
158 will need at least 8MB RAM or more for booting.
Alain Knaff30d65db2009-01-04 22:46:17 +0100159
160config KERNEL_LZMA
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800161 bool "LZMA"
162 depends on HAVE_KERNEL_LZMA
163 help
164 The most recent compression algorithm.
165 Its ratio is best, decompression speed is between the other
166 two. Compression is slowest. The kernel size is about 33%
167 smaller with LZMA in comparison to gzip.
Alain Knaff30d65db2009-01-04 22:46:17 +0100168
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800169config KERNEL_LZO
170 bool "LZO"
171 depends on HAVE_KERNEL_LZO
172 help
173 Its compression ratio is the poorest among the 4. The kernel
174 size is about about 10% bigger than gzip; however its speed
175 (both compression and decompression) is the fastest.
176
Alain Knaff30d65db2009-01-04 22:46:17 +0100177endchoice
178
Linus Torvalds1da177e2005-04-16 15:20:36 -0700179config SWAP
180 bool "Support for paging of anonymous memory (swap)"
David Howells93614012006-09-30 20:45:40 +0200181 depends on MMU && BLOCK
Linus Torvalds1da177e2005-04-16 15:20:36 -0700182 default y
183 help
184 This option allows you to choose whether you want to have support
Jesper Juhl92c35042006-01-15 02:40:08 +0100185 for so called swap devices or swap files in your kernel that are
Linus Torvalds1da177e2005-04-16 15:20:36 -0700186 used to provide more virtual memory than the actual RAM present
187 in your computer. If unsure say Y.
188
189config SYSVIPC
190 bool "System V IPC"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700191 ---help---
192 Inter Process Communication is a suite of library functions and
193 system calls which let processes (running programs) synchronize and
194 exchange information. It is generally considered to be a good thing,
195 and some programs won't run unless you say Y here. In particular, if
196 you want to run the DOS emulator dosemu under Linux (read the
197 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
198 you'll need to say Y here.
199
200 You can find documentation about IPC with "info ipc" and also in
201 section 6.4 of the Linux Programmer's Guide, available from
202 <http://www.tldp.org/guides.html>.
203
Eric W. Biedermana5494dc2007-02-14 00:34:06 -0800204config SYSVIPC_SYSCTL
205 bool
206 depends on SYSVIPC
207 depends on SYSCTL
208 default y
209
Linus Torvalds1da177e2005-04-16 15:20:36 -0700210config POSIX_MQUEUE
211 bool "POSIX Message Queues"
212 depends on NET && EXPERIMENTAL
213 ---help---
214 POSIX variant of message queues is a part of IPC. In POSIX message
215 queues every message has a priority which decides about succession
216 of receiving it by a process. If you want to compile and run
217 programs written e.g. for Solaris with use of its POSIX message
Robert P. J. Dayb0e37652007-05-09 07:25:13 +0200218 queues (functions mq_*) say Y here.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700219
220 POSIX message queues are visible as a filesystem called 'mqueue'
221 and can be mounted somewhere if you want to do filesystem
222 operations on message queues.
223
224 If unsure, say Y.
225
Serge E. Hallynbdc8e5f2009-04-06 19:01:11 -0700226config POSIX_MQUEUE_SYSCTL
227 bool
228 depends on POSIX_MQUEUE
229 depends on SYSCTL
230 default y
231
Linus Torvalds1da177e2005-04-16 15:20:36 -0700232config BSD_PROCESS_ACCT
233 bool "BSD Process Accounting"
234 help
235 If you say Y here, a user level program will be able to instruct the
236 kernel (via a special system call) to write process accounting
237 information to a file: whenever a process exits, information about
238 that process will be appended to the file by the kernel. The
239 information includes things such as creation time, owning user,
240 command name, memory usage, controlling terminal etc. (the complete
241 list is in the struct acct in <file:include/linux/acct.h>). It is
242 up to the user level program to do useful things with this
243 information. This is generally a good idea, so say Y.
244
245config BSD_PROCESS_ACCT_V3
246 bool "BSD Process Accounting version 3 file format"
247 depends on BSD_PROCESS_ACCT
248 default n
249 help
250 If you say Y here, the process accounting information is written
251 in a new file format that also logs the process IDs of each
252 process and it's parent. Note that this file format is incompatible
253 with previous v0/v1/v2 file formats, so you will need updated tools
254 for processing it. A preliminary version of these tools is available
S.Çağlar Onur37a4c942008-06-18 11:45:13 +0300255 at <http://www.gnu.org/software/acct/>.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700256
Shailabh Nagarc7572492006-07-14 00:24:40 -0700257config TASKSTATS
258 bool "Export task/process statistics through netlink (EXPERIMENTAL)"
259 depends on NET
260 default n
261 help
262 Export selected statistics for tasks/processes through the
263 generic netlink interface. Unlike BSD process accounting, the
264 statistics are available during the lifetime of tasks/processes as
265 responses to commands. Like BSD accounting, they are sent to user
266 space on task exit.
267
268 Say N if unsure.
269
Shailabh Nagarca74e922006-07-14 00:24:36 -0700270config TASK_DELAY_ACCT
271 bool "Enable per-task delay accounting (EXPERIMENTAL)"
Shailabh Nagar6f449932006-07-14 00:24:41 -0700272 depends on TASKSTATS
Shailabh Nagarca74e922006-07-14 00:24:36 -0700273 help
274 Collect information on time spent by a task waiting for system
275 resources like cpu, synchronous block I/O completion and swapping
276 in pages. Such statistics can help in setting a task's priorities
277 relative to other tasks for cpu, io, rss limits etc.
278
279 Say N if unsure.
280
Alexey Dobriyan18f705f2007-02-10 01:46:44 -0800281config TASK_XACCT
282 bool "Enable extended accounting over taskstats (EXPERIMENTAL)"
283 depends on TASKSTATS
284 help
285 Collect extended task accounting data and send the data
286 to userland for processing over the taskstats interface.
287
288 Say N if unsure.
289
290config TASK_IO_ACCOUNTING
291 bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
292 depends on TASK_XACCT
293 help
294 Collect information on the number of bytes of storage I/O which this
295 task has caused.
296
297 Say N if unsure.
298
Linus Torvalds1da177e2005-04-16 15:20:36 -0700299config AUDIT
300 bool "Auditing support"
Chris Wright804a6a492005-05-11 10:52:45 +0100301 depends on NET
Linus Torvalds1da177e2005-04-16 15:20:36 -0700302 help
303 Enable auditing infrastructure that can be used with another
304 kernel subsystem, such as SELinux (which requires this for
305 logging of avc messages output). Does not do system-call
306 auditing without CONFIG_AUDITSYSCALL.
307
308config AUDITSYSCALL
309 bool "Enable system-call auditing support"
Kumar Gala022382a2009-10-16 07:21:37 +0000310 depends on AUDIT && (X86 || PPC || S390 || IA64 || UML || SPARC64 || SUPERH)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700311 default y if SECURITY_SELINUX
312 help
313 Enable low-overhead system-call auditing infrastructure that
314 can be used independently or with another kernel subsystem,
Amy Griffisf368c07d2006-04-07 16:55:56 -0400315 such as SELinux. To use audit's filesystem watch feature, please
316 ensure that INOTIFY is configured.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700317
Al Viro74c3cbe2007-07-22 08:04:18 -0400318config AUDIT_TREE
319 def_bool y
Eric Paris63c882a2009-05-21 17:02:01 -0400320 depends on AUDITSYSCALL
321 select INOTIFY
Al Viro74c3cbe2007-07-22 08:04:18 -0400322
Mike Travisc903ff82009-01-15 12:28:29 -0800323menu "RCU Subsystem"
324
325choice
326 prompt "RCU Implementation"
Paul E. McKenney31c9a242009-04-02 21:06:25 -0700327 default TREE_RCU
Mike Travisc903ff82009-01-15 12:28:29 -0800328
Mike Travisc903ff82009-01-15 12:28:29 -0800329config TREE_RCU
330 bool "Tree-based hierarchical RCU"
331 help
332 This option selects the RCU implementation that is
333 designed for very large SMP system with hundreds or
Paul E. McKenneyc17ef452009-06-23 17:12:47 -0700334 thousands of CPUs. It also scales down nicely to
335 smaller systems.
Mike Travisc903ff82009-01-15 12:28:29 -0800336
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700337config TREE_PREEMPT_RCU
338 bool "Preemptable tree-based hierarchical RCU"
339 depends on PREEMPT
340 help
341 This option selects the RCU implementation that is
342 designed for very large SMP systems with hundreds or
343 thousands of CPUs, but for which real-time response
Paul E. McKenneybbe3eae2009-09-13 09:15:08 -0700344 is also required. It also scales down nicely to
345 smaller systems.
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700346
Paul E. McKenney9b1d82f2009-10-25 19:03:50 -0700347config TINY_RCU
348 bool "UP-only small-memory-footprint RCU"
349 depends on !SMP
350 help
351 This option selects the RCU implementation that is
352 designed for UP systems from which real-time response
353 is not required. This option greatly reduces the
354 memory footprint of RCU.
355
Mike Travisc903ff82009-01-15 12:28:29 -0800356endchoice
357
358config RCU_TRACE
359 bool "Enable tracing for RCU"
Paul E. McKenney6b3ef482009-08-22 13:56:53 -0700360 depends on TREE_RCU || TREE_PREEMPT_RCU
Mike Travisc903ff82009-01-15 12:28:29 -0800361 help
362 This option provides tracing in RCU which presents stats
363 in debugfs for debugging RCU implementation.
364
365 Say Y here if you want to enable RCU tracing
366 Say N if you are unsure.
367
368config RCU_FANOUT
369 int "Tree-based hierarchical RCU fanout value"
370 range 2 64 if 64BIT
371 range 2 32 if !64BIT
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700372 depends on TREE_RCU || TREE_PREEMPT_RCU
Mike Travisc903ff82009-01-15 12:28:29 -0800373 default 64 if 64BIT
374 default 32 if !64BIT
375 help
376 This option controls the fanout of hierarchical implementations
377 of RCU, allowing RCU to work efficiently on machines with
378 large numbers of CPUs. This value must be at least the cube
379 root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
380 systems and up to 262,144 for 64-bit systems.
381
382 Select a specific number if testing RCU itself.
383 Take the default if unsure.
384
385config RCU_FANOUT_EXACT
386 bool "Disable tree-based hierarchical RCU auto-balancing"
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700387 depends on TREE_RCU || TREE_PREEMPT_RCU
Mike Travisc903ff82009-01-15 12:28:29 -0800388 default n
389 help
390 This option forces use of the exact RCU_FANOUT value specified,
391 regardless of imbalances in the hierarchy. This is useful for
392 testing RCU itself, and might one day be useful on systems with
393 strong NUMA behavior.
394
395 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
396
397 Say N if unsure.
398
Paul E. McKenney8bd93a22010-02-22 17:04:59 -0800399config RCU_FAST_NO_HZ
400 bool "Accelerate last non-dyntick-idle CPU's grace periods"
401 depends on TREE_RCU && NO_HZ && SMP
402 default n
403 help
404 This option causes RCU to attempt to accelerate grace periods
405 in order to allow the final CPU to enter dynticks-idle state
406 more quickly. On the other hand, this option increases the
407 overhead of the dynticks-idle checking, particularly on systems
408 with large numbers of CPUs.
409
410 Say Y if energy efficiency is critically important, particularly
411 if you have relatively few CPUs.
412
413 Say N if you are unsure.
414
Mike Travisc903ff82009-01-15 12:28:29 -0800415config TREE_RCU_TRACE
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700416 def_bool RCU_TRACE && ( TREE_RCU || TREE_PREEMPT_RCU )
Mike Travisc903ff82009-01-15 12:28:29 -0800417 select DEBUG_FS
418 help
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700419 This option provides tracing for the TREE_RCU and
420 TREE_PREEMPT_RCU implementations, permitting Makefile to
421 trivially select kernel/rcutree_trace.c.
Mike Travisc903ff82009-01-15 12:28:29 -0800422
Mike Travisc903ff82009-01-15 12:28:29 -0800423endmenu # "RCU Subsystem"
424
Linus Torvalds1da177e2005-04-16 15:20:36 -0700425config IKCONFIG
Ross Birof2443ab2006-09-30 23:27:25 -0700426 tristate "Kernel .config support"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700427 ---help---
428 This option enables the complete Linux kernel ".config" file
429 contents to be saved in the kernel. It provides documentation
430 of which kernel options are used in a running kernel or in an
431 on-disk kernel. This information can be extracted from the kernel
432 image file with the script scripts/extract-ikconfig and used as
433 input to rebuild the current kernel or to build another kernel.
434 It can also be extracted from a running kernel by reading
435 /proc/config.gz if enabled (below).
436
437config IKCONFIG_PROC
438 bool "Enable access to .config through /proc/config.gz"
439 depends on IKCONFIG && PROC_FS
440 ---help---
441 This option enables access to the kernel configuration file
442 through /proc/config.gz.
443
Alistair John Strachan794543a2007-05-08 00:31:15 -0700444config LOG_BUF_SHIFT
445 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
446 range 12 21
Adrian Bunkf17a32e2008-04-29 00:58:58 -0700447 default 17
Alistair John Strachan794543a2007-05-08 00:31:15 -0700448 help
449 Select kernel log buffer size as a power of 2.
Adrian Bunkf17a32e2008-04-29 00:58:58 -0700450 Examples:
451 17 => 128 KB
452 16 => 64 KB
453 15 => 32 KB
454 14 => 16 KB
Alistair John Strachan794543a2007-05-08 00:31:15 -0700455 13 => 8 KB
456 12 => 4 KB
457
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800458#
459# Architectures with an unreliable sched_clock() should select this:
460#
461config HAVE_UNSTABLE_SCHED_CLOCK
462 bool
463
464config GROUP_SCHED
465 bool "Group CPU scheduler"
466 depends on EXPERIMENTAL
467 default n
468 help
469 This feature lets CPU scheduler recognize task groups and control CPU
470 bandwidth allocation to such task groups.
471 In order to create a group from arbitrary set of processes, use
472 CONFIG_CGROUPS. (See Control Group support.)
473
474config FAIR_GROUP_SCHED
475 bool "Group scheduling for SCHED_OTHER"
476 depends on GROUP_SCHED
477 default GROUP_SCHED
478
479config RT_GROUP_SCHED
480 bool "Group scheduling for SCHED_RR/FIFO"
481 depends on EXPERIMENTAL
482 depends on GROUP_SCHED
483 default n
484 help
485 This feature lets you explicitly allocate real CPU bandwidth
486 to users or control groups (depending on the "Basis for grouping tasks"
487 setting below. If enabled, it will also make it impossible to
488 schedule realtime tasks for non-root users until you allocate
489 realtime bandwidth for them.
490 See Documentation/scheduler/sched-rt-group.txt for more information.
491
492choice
493 depends on GROUP_SCHED
494 prompt "Basis for grouping tasks"
495 default USER_SCHED
496
497config USER_SCHED
498 bool "user id"
499 help
500 This option will choose userid as the basis for grouping
501 tasks, thus providing equal CPU bandwidth to each user.
502
503config CGROUP_SCHED
504 bool "Control groups"
505 depends on CGROUPS
506 help
507 This option allows you to create arbitrary task groups
508 using the "cgroup" pseudo filesystem and control
509 the cpu bandwidth allocated to each such task group.
Li Zefan45ce80f2009-01-15 13:50:59 -0800510 Refer to Documentation/cgroups/cgroups.txt for more
511 information on "cgroup" pseudo filesystem.
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800512
513endchoice
514
Li Zefan23964d22009-01-15 13:50:58 -0800515menuconfig CGROUPS
516 boolean "Control Group support"
Paul Menageddbcc7e2007-10-18 23:39:30 -0700517 help
Li Zefan23964d22009-01-15 13:50:58 -0800518 This option adds support for grouping sets of processes together, for
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800519 use with process control subsystems such as Cpusets, CFS, memory
520 controls or device isolation.
521 See
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800522 - Documentation/scheduler/sched-design-CFS.txt (CFS)
Li Zefan45ce80f2009-01-15 13:50:59 -0800523 - Documentation/cgroups/ (features for grouping, isolation
524 and resource control)
Paul Menageddbcc7e2007-10-18 23:39:30 -0700525
526 Say N if unsure.
527
Li Zefan23964d22009-01-15 13:50:58 -0800528if CGROUPS
529
Paul Menage006cb992007-10-18 23:39:43 -0700530config CGROUP_DEBUG
531 bool "Example debug cgroup subsystem"
532 depends on CGROUPS
Paul Menage418d7d82008-04-29 01:00:05 -0700533 default n
Paul Menage006cb992007-10-18 23:39:43 -0700534 help
535 This option enables a simple cgroup subsystem that
536 exports useful debugging information about the cgroups
Li Zefan23964d22009-01-15 13:50:58 -0800537 framework.
Paul Menage006cb992007-10-18 23:39:43 -0700538
Li Zefan23964d22009-01-15 13:50:58 -0800539 Say N if unsure.
Paul Menage006cb992007-10-18 23:39:43 -0700540
Serge E. Hallyn858d72e2007-10-18 23:39:45 -0700541config CGROUP_NS
Li Zefan23964d22009-01-15 13:50:58 -0800542 bool "Namespace cgroup subsystem"
543 depends on CGROUPS
544 help
545 Provides a simple namespace cgroup subsystem to
546 provide hierarchical naming of sets of namespaces,
547 for instance virtual servers and checkpoint/restart
548 jobs.
Serge E. Hallyn858d72e2007-10-18 23:39:45 -0700549
Matt Helsleydc52ddc2008-10-18 20:27:21 -0700550config CGROUP_FREEZER
Li Zefan23964d22009-01-15 13:50:58 -0800551 bool "Freezer cgroup subsystem"
552 depends on CGROUPS
553 help
554 Provides a way to freeze and unfreeze all tasks in a
Matt Helsleydc52ddc2008-10-18 20:27:21 -0700555 cgroup.
556
Serge E. Hallyn08ce5f12008-04-29 01:00:10 -0700557config CGROUP_DEVICE
558 bool "Device controller for cgroups"
559 depends on CGROUPS && EXPERIMENTAL
560 help
561 Provides a cgroup implementing whitelists for devices which
562 a process in the cgroup can mknod or open.
563
Linus Torvalds1da177e2005-04-16 15:20:36 -0700564config CPUSETS
565 bool "Cpuset support"
Paul Menagedb7f47c2009-04-02 16:57:55 -0700566 depends on CGROUPS
Linus Torvalds1da177e2005-04-16 15:20:36 -0700567 help
Randy Dunlapd9fd8a62005-07-27 11:45:11 -0700568 This option will let you create and manage CPUSETs which
Linus Torvalds1da177e2005-04-16 15:20:36 -0700569 allow dynamically partitioning a system into sets of CPUs and
570 Memory Nodes and assigning tasks to run only within those sets.
571 This is primarily useful on large SMP or NUMA systems.
572
573 Say N if unsure.
574
Li Zefan23964d22009-01-15 13:50:58 -0800575config PROC_PID_CPUSET
576 bool "Include legacy /proc/<pid>/cpuset file"
577 depends on CPUSETS
578 default y
579
Srivatsa Vaddagirid842de82007-12-02 20:04:49 +0100580config CGROUP_CPUACCT
581 bool "Simple CPU accounting cgroup subsystem"
582 depends on CGROUPS
583 help
584 Provides a simple Resource Controller for monitoring the
Li Zefan23964d22009-01-15 13:50:58 -0800585 total CPU consumed by the tasks in a cgroup.
Srivatsa Vaddagirid842de82007-12-02 20:04:49 +0100586
Pavel Emelianove552b662008-02-07 00:13:49 -0800587config RESOURCE_COUNTERS
588 bool "Resource counters"
589 help
590 This option enables controller independent resource accounting
Li Zefan23964d22009-01-15 13:50:58 -0800591 infrastructure that works with cgroups.
Pavel Emelianove552b662008-02-07 00:13:49 -0800592 depends on CGROUPS
593
Balbir Singh00f0b822008-03-04 14:28:39 -0800594config CGROUP_MEM_RES_CTLR
595 bool "Memory Resource Controller for Control Groups"
596 depends on CGROUPS && RESOURCE_COUNTERS
Balbir Singhcf475ad2008-04-29 01:00:16 -0700597 select MM_OWNER
Balbir Singh00f0b822008-03-04 14:28:39 -0800598 help
KAMEZAWA Hiroyuki84ad6d72008-10-29 14:01:06 -0700599 Provides a memory resource controller that manages both anonymous
Thadeu Lima de Souza Cascardo21acb9c2009-02-04 10:12:08 +0100600 memory and page cache. (See Documentation/cgroups/memory.txt)
Balbir Singh00f0b822008-03-04 14:28:39 -0800601
602 Note that setting this option increases fixed memory overhead
KAMEZAWA Hiroyuki84ad6d72008-10-29 14:01:06 -0700603 associated with each page of memory in the system. By this,
604 20(40)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
605 usage tracking struct at boot. Total amount of this is printed out
606 at boot.
Balbir Singh00f0b822008-03-04 14:28:39 -0800607
608 Only enable when you're ok with these trade offs and really
KAMEZAWA Hiroyuki84ad6d72008-10-29 14:01:06 -0700609 sure you need the memory resource controller. Even when you enable
610 this, you can set "cgroup_disable=memory" at your boot option to
611 disable memory resource controller and you can avoid overheads.
Li Zefanc9d54092009-01-07 18:07:35 -0800612 (and lose benefits of memory resource controller)
Balbir Singh00f0b822008-03-04 14:28:39 -0800613
Balbir Singhcf475ad2008-04-29 01:00:16 -0700614 This config option also selects MM_OWNER config option, which
615 could in turn add some fork/exit overhead.
616
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -0800617config CGROUP_MEM_RES_CTLR_SWAP
618 bool "Memory Resource Controller Swap Extension(EXPERIMENTAL)"
619 depends on CGROUP_MEM_RES_CTLR && SWAP && EXPERIMENTAL
620 help
621 Add swap management feature to memory resource controller. When you
622 enable this, you can limit mem+swap usage per cgroup. In other words,
623 when you disable this, memory resource controller has no cares to
624 usage of swap...a process can exhaust all of the swap. This extension
625 is useful when you want to avoid exhaustion swap but this itself
626 adds more overheads and consumes memory for remembering information.
627 Especially if you use 32bit system or small memory system, please
628 be careful about enabling this. When memory resource controller
629 is disabled by boot option, this will be automatically disabled and
630 there will be no overhead from this. Even when you set this config=y,
631 if boot option "noswapaccount" is set, swap will not be accounted.
KAMEZAWA Hiroyuki627991a2009-04-02 16:57:47 -0700632 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
633 size is 4096bytes, 512k per 1Gbytes of swap.
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -0800634
Li Zefan23964d22009-01-15 13:50:58 -0800635endif # CGROUPS
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -0800636
Li Zefan23964d22009-01-15 13:50:58 -0800637config MM_OWNER
638 bool
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800639
Kay Sievers88a22c92006-09-14 11:23:28 +0200640config SYSFS_DEPRECATED
Ingo Molnard47846c2008-03-04 14:54:47 +0100641 bool
642
643config SYSFS_DEPRECATED_V2
Uwe Kleine-König9e9868a2009-12-03 19:58:00 +0100644 bool "enable deprecated sysfs features to support old userspace tools"
Randy Dunlap9148fe82007-12-31 10:05:34 -0800645 depends on SYSFS
Kay Sieversf6ee6492009-04-16 19:56:37 +0200646 default n
Ingo Molnard47846c2008-03-04 14:54:47 +0100647 select SYSFS_DEPRECATED
Kay Sievers88a22c92006-09-14 11:23:28 +0200648 help
Kay Sieversfce3e802008-11-01 14:03:00 +0100649 This option switches the layout of sysfs to the deprecated
Kay Sieversf6ee6492009-04-16 19:56:37 +0200650 version. Do not use it on recent distributions.
Kay Sievers88a22c92006-09-14 11:23:28 +0200651
Kay Sieversfce3e802008-11-01 14:03:00 +0100652 The current sysfs layout features a unified device tree at
653 /sys/devices/, which is able to express a hierarchy between
654 class devices. If the deprecated option is set to Y, the
655 unified device tree is split into a bus device tree at
656 /sys/devices/ and several individual class device trees at
657 /sys/class/. The class and bus devices will be connected by
658 "<subsystem>:<name>" and the "device" links. The "block"
659 class devices, will not show up in /sys/class/block/. Some
660 subsystems will suppress the creation of some devices which
661 depend on the unified device tree.
Kay Sievers88a22c92006-09-14 11:23:28 +0200662
Kay Sieversfce3e802008-11-01 14:03:00 +0100663 This option is not a pure compatibility option that can
664 be safely enabled on newer distributions. It will change the
665 layout of sysfs to the non-extensible deprecated version,
666 and disable some features, which can not be exported without
667 confusing older userspace tools. Since 2007/2008 all major
668 distributions do not enable this option, and ship no tools which
669 depend on the deprecated layout or this option.
670
671 If you are using a new kernel on an older distribution, or use
672 older userspace tools, you might need to say Y here. Do not say Y,
673 if the original kernel, that came with your distribution, has
674 this option set to N.
Kay Sievers88a22c92006-09-14 11:23:28 +0200675
Jens Axboeb86ff9812006-03-23 19:56:55 +0100676config RELAY
677 bool "Kernel->user space relay support (formerly relayfs)"
678 help
679 This option enables support for relay interface support in
680 certain file systems (such as debugfs).
681 It is designed to provide an efficient mechanism for tools and
682 facilities to relay large amounts of data from kernel space to
683 user space.
684
685 If unsure, say N.
686
Pavel Emelyanovc5289a692008-02-08 04:18:19 -0800687config NAMESPACES
688 bool "Namespaces support" if EMBEDDED
689 default !EMBEDDED
690 help
691 Provides the way to make tasks work with different objects using
692 the same id. For example same IPC id may refer to different objects
693 or same user id or pid may refer to different tasks when used in
694 different namespaces.
695
Pavel Emelyanov58bfdd6d2008-02-08 04:18:21 -0800696config UTS_NS
697 bool "UTS namespace"
698 depends on NAMESPACES
699 help
700 In this namespace tasks see different info provided with the
701 uname() system call
702
Pavel Emelyanovae5e1b22008-02-08 04:18:22 -0800703config IPC_NS
704 bool "IPC namespace"
Serge E. Hallyn614b84c2009-04-06 19:01:08 -0700705 depends on NAMESPACES && (SYSVIPC || POSIX_MQUEUE)
Pavel Emelyanovae5e1b22008-02-08 04:18:22 -0800706 help
707 In this namespace tasks work with IPC ids which correspond to
Serge E. Hallyn614b84c2009-04-06 19:01:08 -0700708 different IPC objects in different namespaces.
Pavel Emelyanovae5e1b22008-02-08 04:18:22 -0800709
Pavel Emelyanovaee16ce2008-02-08 04:18:23 -0800710config USER_NS
711 bool "User namespace (EXPERIMENTAL)"
712 depends on NAMESPACES && EXPERIMENTAL
713 help
714 This allows containers, i.e. vservers, to use user namespaces
715 to provide different user info for different servers.
716 If unsure, say N.
717
Pavel Emelyanov74bd59b2008-02-08 04:18:24 -0800718config PID_NS
719 bool "PID Namespaces (EXPERIMENTAL)"
720 default n
721 depends on NAMESPACES && EXPERIMENTAL
722 help
Heikki Orsila12d2b8f2008-07-06 15:48:02 +0300723 Support process id namespaces. This allows having multiple
Matt LaPlante692105b2009-01-26 11:12:25 +0100724 processes with the same pid as long as they are in different
Pavel Emelyanov74bd59b2008-02-08 04:18:24 -0800725 pid namespaces. This is a building block of containers.
726
727 Unless you want to work with an experimental feature
728 say N here.
729
Matt Helsleyd6eb6332009-01-26 12:25:55 -0800730config NET_NS
731 bool "Network namespace"
732 default n
733 depends on NAMESPACES && EXPERIMENTAL && NET
734 help
735 Allow user space to create what appear to be multiple instances
736 of the network stack.
737
Dimitri Gorokhovikf9916332007-03-06 01:42:17 -0800738config BLK_DEV_INITRD
739 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
740 depends on BROKEN || !FRV
741 help
742 The initial RAM filesystem is a ramfs which is loaded by the
743 boot loader (loadlin or lilo) and that is mounted as root
744 before the normal boot procedure. It is typically used to
745 load modules needed to mount the "real" root file system,
746 etc. See <file:Documentation/initrd.txt> for details.
747
748 If RAM disk support (BLK_DEV_RAM) is also included, this
749 also enables initial RAM disk (initrd) support and adds
750 15 Kbytes (more on some other architectures) to the kernel size.
751
752 If unsure say Y.
753
Jean-Paul Samanc33df4e2007-02-10 01:44:43 -0800754if BLK_DEV_INITRD
755
Sam Ravnborgdbec4862005-08-10 20:44:50 +0200756source "usr/Kconfig"
757
Jean-Paul Samanc33df4e2007-02-10 01:44:43 -0800758endif
759
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800760config CC_OPTIMIZE_FOR_SIZE
Ingo Molnar96fffeb2008-04-28 01:39:43 +0200761 bool "Optimize for size"
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800762 default y
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800763 help
764 Enabling this option will pass "-Os" instead of "-O2" to gcc
765 resulting in a smaller kernel.
766
jkacur775a7222008-07-16 00:31:16 +0200767 If unsure, say Y.
Linus Torvaldsc45b4f12005-12-14 18:52:21 -0800768
Randy Dunlap08470622006-09-30 23:28:13 -0700769config SYSCTL
770 bool
771
Randy Dunlapb943c462009-03-10 12:55:46 -0700772config ANON_INODES
773 bool
774
Linus Torvalds1da177e2005-04-16 15:20:36 -0700775menuconfig EMBEDDED
776 bool "Configure standard kernel features (for small systems)"
777 help
778 This option allows certain base kernel options and settings
779 to be disabled or tweaked. This is for specialized
780 environments which can tolerate a "non-standard" kernel.
781 Only use this if you really know what you are doing.
782
Chuck Ebbertae81f9e2006-09-16 12:15:53 -0700783config UID16
784 bool "Enable 16-bit UID system calls" if EMBEDDED
David S. Miller09337f52008-04-26 03:17:12 -0700785 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 -0700786 default y
787 help
788 This enables the legacy 16-bit UID syscall wrappers.
789
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700790config SYSCTL_SYSCALL
Randy Dunlap08470622006-09-30 23:28:13 -0700791 bool "Sysctl syscall support" if EMBEDDED
Eric W. Biederman26a70342009-11-05 05:26:41 -0800792 depends on PROC_SYSCTL
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800793 default y
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700794 select SYSCTL
795 ---help---
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800796 sys_sysctl uses binary paths that have been found challenging
797 to properly maintain and use. The interface in /proc/sys
798 using paths with ascii names is now the primary path to this
799 information.
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700800
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800801 Almost nothing using the binary sysctl interface so if you are
802 trying to save some space it is probably safe to disable this,
803 making your kernel marginally smaller.
Eric W. Biedermanb89a8172006-09-27 01:51:04 -0700804
Eric W. Biederman13bb7e32006-11-08 17:44:51 -0800805 If unsure say Y here.
Chuck Ebbertae81f9e2006-09-16 12:15:53 -0700806
Linus Torvalds1da177e2005-04-16 15:20:36 -0700807config KALLSYMS
Jesper Juhl979c6a12006-12-12 19:25:11 +0100808 bool "Load all symbols for debugging/ksymoops" if EMBEDDED
Linus Torvalds1da177e2005-04-16 15:20:36 -0700809 default y
810 help
811 Say Y here to let the kernel print out symbolic crash information and
812 symbolic stack backtraces. This increases the size of the kernel
813 somewhat, as all symbols have to be loaded into the kernel image.
814
815config KALLSYMS_ALL
816 bool "Include all symbols in kallsyms"
817 depends on DEBUG_KERNEL && KALLSYMS
818 help
819 Normally kallsyms only contains the symbols of functions, for nicer
820 OOPS messages. Some debuggers can use kallsyms for other
Jesper Juhlf9f97bc2005-07-20 05:43:05 +0200821 symbols too: say Y here to include all symbols, if you need them
822 and you don't care about adding 300k to the size of your kernel.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700823
824 Say N.
825
826config KALLSYMS_EXTRA_PASS
827 bool "Do an extra kallsyms pass"
828 depends on KALLSYMS
829 help
830 If kallsyms is not working correctly, the build will fail with
831 inconsistent kallsyms data. If that occurs, log a bug report and
832 turn on KALLSYMS_EXTRA_PASS which should result in a stable build.
833 Always say N here unless you find a bug in kallsyms, which must be
834 reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while
835 you wait for kallsyms to be fixed.
836
Matt Mackalld59745c2005-05-01 08:59:02 -0700837
Greg Kroah-Hartman712f47c2005-11-16 11:27:07 -0800838config HOTPLUG
839 bool "Support for hot-pluggable devices" if EMBEDDED
840 default y
841 help
842 This option is provided for the case where no hotplug or uevent
843 capabilities is wanted by the kernel. You should only consider
844 disabling this option for embedded systems that do not use modules, a
845 dynamic /dev tree, or dynamic device discovery. Just say Y.
846
Matt Mackalld59745c2005-05-01 08:59:02 -0700847config PRINTK
848 default y
849 bool "Enable support for printk" if EMBEDDED
850 help
851 This option enables normal printk support. Removing it
852 eliminates most of the message strings from the kernel image
853 and makes the kernel more or less silent. As this makes it
854 very difficult to diagnose system problems, saying N here is
855 strongly discouraged.
856
Matt Mackallc8538a72005-05-01 08:59:01 -0700857config BUG
858 bool "BUG() support" if EMBEDDED
859 default y
860 help
861 Disabling this option eliminates support for BUG and WARN, reducing
862 the size of your kernel image and potentially quietly ignoring
863 numerous fatal conditions. You should only consider disabling this
864 option for embedded systems with no facilities for reporting errors.
865 Just say Y.
866
Matt Mackall708e9a72006-01-08 01:05:25 -0800867config ELF_CORE
868 default y
869 bool "Enable ELF core dumps" if EMBEDDED
870 help
871 Enable support for generating core dumps. Disabling saves about 4k.
872
Stas Sergeeve5e1d3c2008-05-07 12:39:56 +0200873config PCSPKR_PLATFORM
874 bool "Enable PC-Speaker support" if EMBEDDED
875 depends on ALPHA || X86 || MIPS || PPC_PREP || PPC_CHRP || PPC_PSERIES
876 default y
877 help
878 This option allows to disable the internal PC-Speaker
879 support, saving some memory.
880
Linus Torvalds1da177e2005-04-16 15:20:36 -0700881config BASE_FULL
882 default y
883 bool "Enable full-sized data structures for core" if EMBEDDED
884 help
885 Disabling this option reduces the size of miscellaneous core
886 kernel data structures. This saves memory on small machines,
887 but may reduce performance.
888
889config FUTEX
890 bool "Enable futex support" if EMBEDDED
891 default y
Ingo Molnar23f78d4a2006-06-27 02:54:53 -0700892 select RT_MUTEXES
Linus Torvalds1da177e2005-04-16 15:20:36 -0700893 help
894 Disabling this option will cause the kernel to be built without
895 support for "fast userspace mutexes". The resulting kernel may not
896 run glibc-based applications correctly.
897
898config EPOLL
899 bool "Enable eventpoll support" if EMBEDDED
900 default y
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700901 select ANON_INODES
Linus Torvalds1da177e2005-04-16 15:20:36 -0700902 help
903 Disabling this option will cause the kernel to be built without
904 support for epoll family of system calls.
905
Davide Libenzifba2afa2007-05-10 22:23:13 -0700906config SIGNALFD
907 bool "Enable signalfd() system call" if EMBEDDED
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700908 select ANON_INODES
Davide Libenzifba2afa2007-05-10 22:23:13 -0700909 default y
910 help
911 Enable the signalfd() system call that allows to receive signals
912 on a file descriptor.
913
914 If unsure, say Y.
915
Davide Libenzib215e282007-05-10 22:23:16 -0700916config TIMERFD
917 bool "Enable timerfd() system call" if EMBEDDED
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700918 select ANON_INODES
Davide Libenzib215e282007-05-10 22:23:16 -0700919 default y
920 help
921 Enable the timerfd() system call that allows to receive timer
922 events on a file descriptor.
923
924 If unsure, say Y.
925
Davide Libenzie1ad7462007-05-10 22:23:19 -0700926config EVENTFD
927 bool "Enable eventfd() system call" if EMBEDDED
Adrian Bunk448e3ce2007-07-31 00:39:10 -0700928 select ANON_INODES
Davide Libenzie1ad7462007-05-10 22:23:19 -0700929 default y
930 help
931 Enable the eventfd() system call that allows to receive both
932 kernel notification (ie. KAIO) or userspace notifications.
933
934 If unsure, say Y.
935
Linus Torvalds1da177e2005-04-16 15:20:36 -0700936config SHMEM
937 bool "Use full shmem filesystem" if EMBEDDED
938 default y
939 depends on MMU
940 help
941 The shmem is an internal filesystem used to manage shared memory.
942 It is backed by swap and manages resource limits. It is also exported
943 to userspace as tmpfs if TMPFS is enabled. Disabling this
944 option replaces shmem and tmpfs with the much simpler ramfs code,
945 which may be appropriate on small systems without swap.
946
Thomas Petazzoniebf3f092008-10-15 22:05:12 -0700947config AIO
948 bool "Enable AIO support" if EMBEDDED
949 default y
950 help
951 This option enables POSIX asynchronous I/O which may by used
952 by some high performance threaded applications. Disabling
953 this option saves about 7k.
954
Ingo Molnarcdd6c482009-09-21 12:02:48 +0200955config HAVE_PERF_EVENTS
Thomas Gleixner0793a612008-12-04 20:12:29 +0100956 bool
Mike Frysinger018df722009-06-12 13:17:43 -0400957 help
958 See tools/perf/design.txt for details.
Thomas Gleixner0793a612008-12-04 20:12:29 +0100959
Peter Zijlstra906010b2009-09-21 16:08:49 +0200960config PERF_USE_VMALLOC
961 bool
962 help
963 See tools/perf/design.txt for details
964
Ingo Molnar57c0c152009-09-21 12:20:38 +0200965menu "Kernel Performance Events And Counters"
Thomas Gleixner0793a612008-12-04 20:12:29 +0100966
Ingo Molnarcdd6c482009-09-21 12:02:48 +0200967config PERF_EVENTS
Ingo Molnar57c0c152009-09-21 12:20:38 +0200968 bool "Kernel performance events and counters"
969 default y if (PROFILING || PERF_COUNTERS)
Ingo Molnarcdd6c482009-09-21 12:02:48 +0200970 depends on HAVE_PERF_EVENTS
Ingo Molnar4c59e462008-12-08 19:38:33 +0100971 select ANON_INODES
Thomas Gleixner0793a612008-12-04 20:12:29 +0100972 help
Ingo Molnar57c0c152009-09-21 12:20:38 +0200973 Enable kernel support for various performance events provided
974 by software and hardware.
Thomas Gleixner0793a612008-12-04 20:12:29 +0100975
Thadeu Lima de Souza Cascardodd770382009-10-30 19:32:25 -0200976 Software events are supported either built-in or via the
Ingo Molnar57c0c152009-09-21 12:20:38 +0200977 use of generic tracepoints.
978
979 Most modern CPUs support performance events via performance
980 counter registers. These registers count the number of certain
Thomas Gleixner0793a612008-12-04 20:12:29 +0100981 types of hw events: such as instructions executed, cachemisses
982 suffered, or branches mis-predicted - without slowing down the
983 kernel or applications. These registers can also trigger interrupts
984 when a threshold number of events have passed - and can thus be
985 used to profile the code that runs on that CPU.
986
Ingo Molnar57c0c152009-09-21 12:20:38 +0200987 The Linux Performance Event subsystem provides an abstraction of
Thadeu Lima de Souza Cascardodd770382009-10-30 19:32:25 -0200988 these software and hardware event capabilities, available via a
Ingo Molnar57c0c152009-09-21 12:20:38 +0200989 system call and used by the "perf" utility in tools/perf/. It
Thomas Gleixner0793a612008-12-04 20:12:29 +0100990 provides per task and per CPU counters, and it provides event
991 capabilities on top of those.
992
993 Say Y if unsure.
994
Peter Zijlstrae077df42009-03-19 20:26:17 +0100995config EVENT_PROFILE
Peter Zijlstra470a1392009-07-29 10:50:09 +0200996 bool "Tracepoint profiling sources"
Ingo Molnarcdd6c482009-09-21 12:02:48 +0200997 depends on PERF_EVENTS && EVENT_TRACING
Peter Zijlstrae077df42009-03-19 20:26:17 +0100998 default y
Peter Zijlstra470a1392009-07-29 10:50:09 +0200999 help
Ingo Molnar57c0c152009-09-21 12:20:38 +02001000 Allow the use of tracepoints as software performance events.
Peter Zijlstra470a1392009-07-29 10:50:09 +02001001
Ingo Molnar57c0c152009-09-21 12:20:38 +02001002 When this is enabled, you can create perf events based on
Peter Zijlstra470a1392009-07-29 10:50:09 +02001003 tracepoints using PERF_TYPE_TRACEPOINT and the tracepoint ID
1004 found in debugfs://tracing/events/*/*/id. (The -e/--events
1005 option to the perf tool can parse and interpret symbolic
1006 tracepoints, in the subsystem:tracepoint_name format.)
Peter Zijlstrae077df42009-03-19 20:26:17 +01001007
Ingo Molnar57c0c152009-09-21 12:20:38 +02001008config PERF_COUNTERS
1009 bool "Kernel performance counters (old config option)"
1010 depends on HAVE_PERF_EVENTS
1011 help
1012 This config has been obsoleted by the PERF_EVENTS
1013 config option - please see that one for details.
1014
1015 It has no effect on the kernel whether you enable
1016 it or not, it is a compatibility placeholder.
1017
1018 Say N if unsure.
1019
Peter Zijlstra906010b2009-09-21 16:08:49 +02001020config DEBUG_PERF_USE_VMALLOC
1021 default n
1022 bool "Debug: use vmalloc to back perf mmap() buffers"
1023 depends on PERF_EVENTS && DEBUG_KERNEL
1024 select PERF_USE_VMALLOC
1025 help
1026 Use vmalloc memory to back perf mmap() buffers.
1027
1028 Mostly useful for debugging the vmalloc code on platforms
1029 that don't require it.
1030
1031 Say N if unsure.
1032
Thomas Gleixner0793a612008-12-04 20:12:29 +01001033endmenu
1034
Christoph Lameterf8891e52006-06-30 01:55:45 -07001035config VM_EVENT_COUNTERS
1036 default y
1037 bool "Enable VM event counters for /proc/vmstat" if EMBEDDED
1038 help
Paul Jackson2aea4fb2006-12-22 01:06:10 -08001039 VM event counters are needed for event counts to be shown.
1040 This option allows the disabling of the VM event counters
1041 on EMBEDDED systems. /proc/vmstat will only show page counts
1042 if VM event counters are disabled.
Christoph Lameterf8891e52006-06-30 01:55:45 -07001043
Thomas Petazzoni3d137312008-08-19 10:28:24 +02001044config PCI_QUIRKS
1045 default y
Geert Uytterhoeven61cfc7e2008-10-22 08:53:25 +02001046 bool "Enable PCI quirk workarounds" if EMBEDDED
1047 depends on PCI
Thomas Petazzoni3d137312008-08-19 10:28:24 +02001048 help
1049 This enables workarounds for various PCI chipset
1050 bugs/quirks. Disable this only if your target machine is
1051 unaffected by PCI quirks.
1052
Christoph Lameter41ecc552007-05-09 02:32:44 -07001053config SLUB_DEBUG
1054 default y
1055 bool "Enable SLUB debugging support" if EMBEDDED
Christoph Lameterf6acb632008-04-29 16:16:06 -07001056 depends on SLUB && SYSFS
Christoph Lameter41ecc552007-05-09 02:32:44 -07001057 help
1058 SLUB has extensive debug support features. Disabling these can
1059 result in significant savings in code size. This also disables
1060 SLUB sysfs support. /sys/slab will not exist and there will be
1061 no support for cache validation etc.
1062
Randy Dunlapb943c462009-03-10 12:55:46 -07001063config COMPAT_BRK
1064 bool "Disable heap randomization"
1065 default y
1066 help
1067 Randomizing heap placement makes heap exploits harder, but it
1068 also breaks ancient binaries (including anything libc5 based).
1069 This option changes the bootup default to heap randomization
Matt LaPlante692105b2009-01-26 11:12:25 +01001070 disabled, and can be overridden at runtime by setting
Randy Dunlapb943c462009-03-10 12:55:46 -07001071 /proc/sys/kernel/randomize_va_space to 2.
1072
1073 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1074
Christoph Lameter81819f02007-05-06 14:49:36 -07001075choice
1076 prompt "Choose SLAB allocator"
Christoph Lametera0acd822007-07-17 04:03:32 -07001077 default SLUB
Christoph Lameter81819f02007-05-06 14:49:36 -07001078 help
1079 This option allows to select a slab allocator.
1080
1081config SLAB
1082 bool "SLAB"
1083 help
1084 The regular slab allocator that is established and known to work
Christoph Lameter34013882007-05-09 02:32:47 -07001085 well in all environments. It organizes cache hot objects in
Simon Arlott02f56212008-11-05 22:18:19 +00001086 per cpu and per node queues.
Christoph Lameter81819f02007-05-06 14:49:36 -07001087
1088config SLUB
Christoph Lameter81819f02007-05-06 14:49:36 -07001089 bool "SLUB (Unqueued Allocator)"
1090 help
1091 SLUB is a slab allocator that minimizes cache line usage
1092 instead of managing queues of cached objects (SLAB approach).
1093 Per cpu caching is realized using slabs of objects instead
1094 of queues of objects. SLUB can use memory efficiently
Simon Arlott02f56212008-11-05 22:18:19 +00001095 and has enhanced diagnostics. SLUB is the default choice for
1096 a slab allocator.
Christoph Lameter81819f02007-05-06 14:49:36 -07001097
1098config SLOB
Paul Mundt84a01c22007-07-15 23:38:24 -07001099 depends on EMBEDDED
Christoph Lameter81819f02007-05-06 14:49:36 -07001100 bool "SLOB (Simple Allocator)"
1101 help
Matt Mackall37291452008-02-04 22:29:38 -08001102 SLOB replaces the stock allocator with a drastically simpler
1103 allocator. SLOB is generally more space efficient but
1104 does not perform as well on large systems.
Christoph Lameter81819f02007-05-06 14:49:36 -07001105
1106endchoice
1107
Jie Zhangea637632009-12-14 18:00:02 -08001108config MMAP_ALLOW_UNINITIALIZED
1109 bool "Allow mmapped anonymous memory to be uninitialized"
1110 depends on EMBEDDED && !MMU
1111 default n
1112 help
1113 Normally, and according to the Linux spec, anonymous memory obtained
1114 from mmap() has it's contents cleared before it is passed to
1115 userspace. Enabling this config option allows you to request that
1116 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1117 providing a huge performance boost. If this option is not enabled,
1118 then the flag will be ignored.
1119
1120 This is taken advantage of by uClibc's malloc(), and also by
1121 ELF-FDPIC binfmt's brk and stack allocator.
1122
1123 Because of the obvious security issues, this option should only be
1124 enabled on embedded devices where you control what is run in
1125 userspace. Since that isn't generally a problem on no-MMU systems,
1126 it is normally safe to say Y here.
1127
1128 See Documentation/nommu-mmap.txt for more information.
1129
Mathieu Desnoyers125e5642008-02-02 15:10:36 -05001130config PROFILING
1131 bool "Profiling support (EXPERIMENTAL)"
1132 help
1133 Say Y here to enable the extended profiling support mechanisms used
1134 by profilers such as OProfile.
1135
Ingo Molnar5f87f112008-07-23 14:15:22 +02001136#
1137# Place an empty function call at each tracepoint site. Can be
1138# dynamically changed for a probe function.
1139#
Mathieu Desnoyers97e1c182008-07-18 12:16:16 -04001140config TRACEPOINTS
Ingo Molnar5f87f112008-07-23 14:15:22 +02001141 bool
Mathieu Desnoyers97e1c182008-07-18 12:16:16 -04001142
Mathieu Desnoyersfb32e032008-02-02 15:10:33 -05001143source "arch/Kconfig"
1144
David Howells07fe7cb2009-04-03 16:42:35 +01001145config SLOW_WORK
1146 default n
David Howells1c2d0082009-04-06 15:47:25 +01001147 bool
David Howells07fe7cb2009-04-03 16:42:35 +01001148 help
1149 The slow work thread pool provides a number of dynamically allocated
1150 threads that can be used by the kernel to perform operations that
1151 take a relatively long time.
1152
1153 An example of this would be CacheFiles doing a path lookup followed
1154 by a series of mkdirs and a create call, all of which have to touch
1155 disk.
1156
David Howells1c2d0082009-04-06 15:47:25 +01001157 See Documentation/slow-work.txt.
1158
David Howellsf13a48b2009-12-01 15:36:11 +00001159config SLOW_WORK_DEBUG
1160 bool "Slow work debugging through debugfs"
David Howells8fba10a2009-11-19 18:10:51 +00001161 default n
David Howellsf13a48b2009-12-01 15:36:11 +00001162 depends on SLOW_WORK && DEBUG_FS
David Howells8fba10a2009-11-19 18:10:51 +00001163 help
David Howellsf13a48b2009-12-01 15:36:11 +00001164 Display the contents of the slow work run queue through debugfs,
David Howells8fba10a2009-11-19 18:10:51 +00001165 including items currently executing.
1166
1167 See Documentation/slow-work.txt.
1168
Linus Torvalds1da177e2005-04-16 15:20:36 -07001169endmenu # General setup
1170
Dmitry Baryshkovee7e5512008-06-29 14:18:46 +04001171config HAVE_GENERIC_DMA_COHERENT
1172 bool
1173 default n
1174
Linus Torvalds158a9622008-01-02 13:04:48 -08001175config SLABINFO
1176 bool
1177 depends on PROC_FS
Christoph Lameter0f389ec2008-04-14 18:53:02 +03001178 depends on SLAB || SLUB_DEBUG
Linus Torvalds158a9622008-01-02 13:04:48 -08001179 default y
1180
Chuck Ebbertae81f9e2006-09-16 12:15:53 -07001181config RT_MUTEXES
1182 boolean
Chuck Ebbertae81f9e2006-09-16 12:15:53 -07001183
Linus Torvalds1da177e2005-04-16 15:20:36 -07001184config BASE_SMALL
1185 int
1186 default 0 if BASE_FULL
1187 default 1 if !BASE_FULL
1188
Jan Engelhardt66da5732007-07-15 23:39:29 -07001189menuconfig MODULES
Linus Torvalds1da177e2005-04-16 15:20:36 -07001190 bool "Enable loadable module support"
1191 help
1192 Kernel modules are small pieces of compiled code which can
1193 be inserted in the running kernel, rather than being
1194 permanently built into the kernel. You use the "modprobe"
1195 tool to add (and sometimes remove) them. If you say Y here,
1196 many parts of the kernel can be built as modules (by
1197 answering M instead of Y where indicated): this is most
1198 useful for infrequently used options which are not required
1199 for booting. For more information, see the man pages for
1200 modprobe, lsmod, modinfo, insmod and rmmod.
1201
1202 If you say Y here, you will need to run "make
1203 modules_install" to put the modules under /lib/modules/
1204 where modprobe can find them (you may need to be root to do
1205 this).
1206
1207 If unsure, say Y.
1208
Robert P. J. Day0b0de142008-08-04 13:31:32 -04001209if MODULES
1210
Linus Torvalds826e4502008-05-04 17:04:16 -07001211config MODULE_FORCE_LOAD
1212 bool "Forced module loading"
Linus Torvalds826e4502008-05-04 17:04:16 -07001213 default n
1214 help
Rusty Russell91e37a72008-05-09 16:25:28 +10001215 Allow loading of modules without version information (ie. modprobe
1216 --force). Forced module loading sets the 'F' (forced) taint flag and
1217 is usually a really bad idea.
Linus Torvalds826e4502008-05-04 17:04:16 -07001218
Linus Torvalds1da177e2005-04-16 15:20:36 -07001219config MODULE_UNLOAD
1220 bool "Module unloading"
Linus Torvalds1da177e2005-04-16 15:20:36 -07001221 help
1222 Without this option you will not be able to unload any
1223 modules (note that some modules may not be unloadable
Denys Vlasenkof7f5b672008-07-22 19:24:26 -05001224 anyway), which makes your kernel smaller, faster
1225 and simpler. If unsure, say Y.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001226
1227config MODULE_FORCE_UNLOAD
1228 bool "Forced module unloading"
1229 depends on MODULE_UNLOAD && EXPERIMENTAL
1230 help
1231 This option allows you to force a module to unload, even if the
1232 kernel believes it is unsafe: the kernel will remove the module
1233 without waiting for anyone to stop using it (using the -f option to
1234 rmmod). This is mainly for kernel developers and desperate users.
1235 If unsure, say N.
1236
Linus Torvalds1da177e2005-04-16 15:20:36 -07001237config MODVERSIONS
Sam Ravnborg0d541642005-12-26 23:04:02 +01001238 bool "Module versioning support"
Linus Torvalds1da177e2005-04-16 15:20:36 -07001239 help
1240 Usually, you have to use modules compiled with your kernel.
1241 Saying Y here makes it sometimes possible to use modules
1242 compiled for different kernels, by adding enough information
1243 to the modules to (hopefully) spot any changes which would
1244 make them incompatible with the kernel you are running. If
1245 unsure, say N.
1246
1247config MODULE_SRCVERSION_ALL
1248 bool "Source checksum for all modules"
Linus Torvalds1da177e2005-04-16 15:20:36 -07001249 help
1250 Modules which contain a MODULE_VERSION get an extra "srcversion"
1251 field inserted into their modinfo section, which contains a
1252 sum of the source files which made it. This helps maintainers
1253 see exactly which source was used to build a module (since
1254 others sometimes change the module source without updating
1255 the version). With this option, such a "srcversion" field
1256 will be created for all modules. If unsure, say N.
1257
Robert P. J. Day0b0de142008-08-04 13:31:32 -04001258endif # MODULES
1259
Rusty Russell98a79d62008-12-13 21:19:41 +10301260config INIT_ALL_POSSIBLE
1261 bool
1262 help
1263 Back when each arch used to define their own cpu_online_map and
1264 cpu_possible_map, some of them chose to initialize cpu_possible_map
1265 with all 1s, and others with all 0s. When they were centralised,
1266 it was better to provide this option than to break all the archs
Matt LaPlante692105b2009-01-26 11:12:25 +01001267 and have several arch maintainers pursuing me down dark alleys.
Rusty Russell98a79d62008-12-13 21:19:41 +10301268
Linus Torvalds1da177e2005-04-16 15:20:36 -07001269config STOP_MACHINE
1270 bool
1271 default y
1272 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
1273 help
1274 Need stop_machine() primitive.
Jens Axboe3a65dfe2005-11-04 08:43:35 +01001275
Jens Axboe3a65dfe2005-11-04 08:43:35 +01001276source "block/Kconfig"
Avi Kivitye98c3202007-10-16 23:27:31 -07001277
1278config PREEMPT_NOTIFIERS
1279 bool
Paul E. McKenneye260be62008-01-25 21:08:24 +01001280
Thomas Gleixner6beb0002009-11-09 15:21:34 +00001281source "kernel/Kconfig.locks"