blob: 79c6aad5ea15221d89e765b018e5420951fd3548 [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
Peter Oberparleiterb99b87f72009-06-17 16:28:03 -070022
Peter Zijlstrae360adb2010-10-14 14:01:34 +080023config IRQ_WORK
24 bool
Peter Zijlstrae360adb2010-10-14 14:01:34 +080025
David Daney1dbdc6f2012-04-19 14:59:57 -070026config BUILDTIME_EXTABLE_SORT
27 bool
28
Al Boldiff0cfc62007-07-31 00:39:23 -070029menu "General setup"
Linus Torvalds1da177e2005-04-16 15:20:36 -070030
Linus Torvalds1da177e2005-04-16 15:20:36 -070031config BROKEN
32 bool
Linus Torvalds1da177e2005-04-16 15:20:36 -070033
34config BROKEN_ON_SMP
35 bool
36 depends on BROKEN || !SMP
37 default y
38
Linus Torvalds1da177e2005-04-16 15:20:36 -070039config INIT_ENV_ARG_LIMIT
40 int
Adrian Bunkdd673bc2006-06-30 01:55:51 -070041 default 32 if !UML
42 default 128 if UML
Linus Torvalds1da177e2005-04-16 15:20:36 -070043 help
Randy Dunlap34ad92c22005-10-30 15:01:46 -080044 Maximum of each of the number of arguments and environment
45 variables passed to init from the kernel command line.
Linus Torvalds1da177e2005-04-16 15:20:36 -070046
Linus Torvalds1da177e2005-04-16 15:20:36 -070047
Roland McGrath84336462009-12-21 16:24:06 -080048config CROSS_COMPILE
49 string "Cross-compiler tool prefix"
50 help
51 Same as running 'make CROSS_COMPILE=prefix-' but stored for
52 default make runs in this kernel build directory. You don't
53 need to set this unless you want the configured kernel build
54 directory to select the cross-compiler automatically.
55
Jiri Slaby4bb16672013-05-22 10:56:24 +020056config COMPILE_TEST
57 bool "Compile also drivers which will not load"
58 default n
59 help
60 Some drivers can be compiled on a different platform than they are
61 intended to be run on. Despite they cannot be loaded there (or even
62 when they load they cannot be used due to missing HW support),
63 developers still, opposing to distributors, might want to build such
64 drivers to compile-test them.
65
66 If you are a developer and want to build everything available, say Y
67 here. If you are a user/distributor, say N here to exclude useless
68 drivers to be distributed.
69
Linus Torvalds1da177e2005-04-16 15:20:36 -070070config LOCALVERSION
71 string "Local version - append to kernel release"
72 help
73 Append an extra string to the end of your kernel version.
74 This will show up when you type uname, for example.
75 The string you set here will be appended after the contents of
76 any files with a filename matching localversion* in your
77 object and source tree, in that order. Your total string can
78 be a maximum of 64 characters.
79
Ryan Andersonaaebf432005-07-31 04:57:49 -040080config LOCALVERSION_AUTO
81 bool "Automatically append version information to the version string"
82 default y
83 help
84 This will try to automatically determine if the current tree is a
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020085 release tree by looking for git tags that belong to the current
86 top of tree revision.
Ryan Andersonaaebf432005-07-31 04:57:49 -040087
88 A string of the format -gxxxxxxxx will be added to the localversion
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020089 if a git-based tree is found. The string generated by this will be
Ryan Andersonaaebf432005-07-31 04:57:49 -040090 appended after any matching localversion* files, and after the value
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020091 set in CONFIG_LOCALVERSION.
Ryan Andersonaaebf432005-07-31 04:57:49 -040092
Robert P. J. Day6e5a5422007-05-01 23:08:11 +020093 (The actual string used here is the first eight characters produced
94 by running the command:
95
96 $ git rev-parse --verify HEAD
97
98 which is done within the script "scripts/setlocalversion".)
Ryan Andersonaaebf432005-07-31 04:57:49 -040099
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800100config HAVE_KERNEL_GZIP
101 bool
102
103config HAVE_KERNEL_BZIP2
104 bool
105
106config HAVE_KERNEL_LZMA
107 bool
108
Lasse Collin3ebe1242011-01-12 17:01:23 -0800109config HAVE_KERNEL_XZ
110 bool
111
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800112config HAVE_KERNEL_LZO
113 bool
114
Kyungsik Leee76e1fd2013-07-08 16:01:46 -0700115config HAVE_KERNEL_LZ4
116 bool
117
Alain Knaff30d65db2009-01-04 22:46:17 +0100118choice
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800119 prompt "Kernel compression mode"
120 default KERNEL_GZIP
H. Peter Anvin2d3c6272013-11-14 21:43:47 -0800121 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800122 help
Alain Knaff30d65db2009-01-04 22:46:17 +0100123 The linux kernel is a kind of self-extracting executable.
124 Several compression algorithms are available, which differ
125 in efficiency, compression and decompression speed.
126 Compression speed is only relevant when building a kernel.
127 Decompression speed is relevant at each boot.
128
129 If you have any problems with bzip2 or lzma compressed
130 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
131 version of this functionality (bzip2 only), for 2.4, was
132 supplied by Christian Ludwig)
133
134 High compression options are mostly useful for users, who
135 are low on disk space (embedded systems), but for whom ram
136 size matters less.
137
138 If in doubt, select 'gzip'
139
140config KERNEL_GZIP
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800141 bool "Gzip"
142 depends on HAVE_KERNEL_GZIP
143 help
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800144 The old and tried gzip compression. It provides a good balance
145 between compression ratio and decompression speed.
Alain Knaff30d65db2009-01-04 22:46:17 +0100146
147config KERNEL_BZIP2
148 bool "Bzip2"
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800149 depends on HAVE_KERNEL_BZIP2
Alain Knaff30d65db2009-01-04 22:46:17 +0100150 help
151 Its compression ratio and speed is intermediate.
Randy Dunlap0a4dd352012-05-31 16:26:46 -0700152 Decompression speed is slowest among the choices. The kernel
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800153 size is about 10% smaller with bzip2, in comparison to gzip.
154 Bzip2 uses a large amount of memory. For modern kernels you
155 will need at least 8MB RAM or more for booting.
Alain Knaff30d65db2009-01-04 22:46:17 +0100156
157config KERNEL_LZMA
H. Peter Anvin2e9f3bd2009-01-04 15:41:25 -0800158 bool "LZMA"
159 depends on HAVE_KERNEL_LZMA
160 help
Randy Dunlap0a4dd352012-05-31 16:26:46 -0700161 This compression algorithm's ratio is best. Decompression speed
162 is between gzip and bzip2. Compression is slowest.
163 The kernel size is about 33% smaller with LZMA in comparison to gzip.
Alain Knaff30d65db2009-01-04 22:46:17 +0100164
Lasse Collin3ebe1242011-01-12 17:01:23 -0800165config KERNEL_XZ
166 bool "XZ"
167 depends on HAVE_KERNEL_XZ
168 help
169 XZ uses the LZMA2 algorithm and instruction set specific
170 BCJ filters which can improve compression ratio of executable
171 code. The size of the kernel is about 30% smaller with XZ in
172 comparison to gzip. On architectures for which there is a BCJ
173 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
174 will create a few percent smaller kernel than plain LZMA.
175
176 The speed is about the same as with LZMA: The decompression
177 speed of XZ is better than that of bzip2 but worse than gzip
178 and LZO. Compression is slow.
179
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800180config KERNEL_LZO
181 bool "LZO"
182 depends on HAVE_KERNEL_LZO
183 help
Randy Dunlap0a4dd352012-05-31 16:26:46 -0700184 Its compression ratio is the poorest among the choices. The kernel
Stephan Sperber681b3042010-07-14 11:23:08 +0200185 size is about 10% bigger than gzip; however its speed
Albin Tonnerre7dd65fe2010-01-08 14:42:42 -0800186 (both compression and decompression) is the fastest.
187
Kyungsik Leee76e1fd2013-07-08 16:01:46 -0700188config KERNEL_LZ4
189 bool "LZ4"
190 depends on HAVE_KERNEL_LZ4
191 help
192 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
193 A preliminary version of LZ4 de/compression tool is available at
194 <https://code.google.com/p/lz4/>.
195
196 Its compression ratio is worse than LZO. The size of the kernel
197 is about 8% bigger than LZO. But the decompression speed is
198 faster than LZO.
199
Alain Knaff30d65db2009-01-04 22:46:17 +0100200endchoice
201
Josh Triplettbd5dc172011-06-15 15:08:28 -0700202config DEFAULT_HOSTNAME
203 string "Default hostname"
204 default "(none)"
205 help
206 This option determines the default system hostname before userspace
207 calls sethostname(2). The kernel traditionally uses "(none)" here,
208 but you may wish to use a different default here to make a minimal
209 system more usable with less configuration.
210
Linus Torvalds1da177e2005-04-16 15:20:36 -0700211config SWAP
212 bool "Support for paging of anonymous memory (swap)"
David Howells93614012006-09-30 20:45:40 +0200213 depends on MMU && BLOCK
Linus Torvalds1da177e2005-04-16 15:20:36 -0700214 default y
215 help
216 This option allows you to choose whether you want to have support
Jesper Juhl92c35042006-01-15 02:40:08 +0100217 for so called swap devices or swap files in your kernel that are
Linus Torvalds1da177e2005-04-16 15:20:36 -0700218 used to provide more virtual memory than the actual RAM present
219 in your computer. If unsure say Y.
220
221config SYSVIPC
222 bool "System V IPC"
Linus Torvalds1da177e2005-04-16 15:20:36 -0700223 ---help---
224 Inter Process Communication is a suite of library functions and
225 system calls which let processes (running programs) synchronize and
226 exchange information. It is generally considered to be a good thing,
227 and some programs won't run unless you say Y here. In particular, if
228 you want to run the DOS emulator dosemu under Linux (read the
229 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
230 you'll need to say Y here.
231
232 You can find documentation about IPC with "info ipc" and also in
233 section 6.4 of the Linux Programmer's Guide, available from
234 <http://www.tldp.org/guides.html>.
235
Eric W. Biedermana5494dc2007-02-14 00:34:06 -0800236config SYSVIPC_SYSCTL
237 bool
238 depends on SYSVIPC
239 depends on SYSCTL
240 default y
241
Linus Torvalds1da177e2005-04-16 15:20:36 -0700242config POSIX_MQUEUE
243 bool "POSIX Message Queues"
Kees Cook19c92392012-10-02 11:19:29 -0700244 depends on NET
Linus Torvalds1da177e2005-04-16 15:20:36 -0700245 ---help---
246 POSIX variant of message queues is a part of IPC. In POSIX message
247 queues every message has a priority which decides about succession
248 of receiving it by a process. If you want to compile and run
249 programs written e.g. for Solaris with use of its POSIX message
Robert P. J. Dayb0e37652007-05-09 07:25:13 +0200250 queues (functions mq_*) say Y here.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700251
252 POSIX message queues are visible as a filesystem called 'mqueue'
253 and can be mounted somewhere if you want to do filesystem
254 operations on message queues.
255
256 If unsure, say Y.
257
Serge E. Hallynbdc8e5f2009-04-06 19:01:11 -0700258config POSIX_MQUEUE_SYSCTL
259 bool
260 depends on POSIX_MQUEUE
261 depends on SYSCTL
262 default y
263
Konstantin Khlebnikov226b4cc2014-06-04 16:10:50 -0700264config CROSS_MEMORY_ATTACH
265 bool "Enable process_vm_readv/writev syscalls"
266 depends on MMU
267 default y
268 help
269 Enabling this option adds the system calls process_vm_readv and
270 process_vm_writev which allow a process with the correct privileges
Geert Uytterhoevena2a368d2014-08-12 13:46:11 -0700271 to directly read from or write to another process' address space.
Konstantin Khlebnikov226b4cc2014-06-04 16:10:50 -0700272 See the man page for more details.
273
Aneesh Kumar K.V990d6c22011-01-29 18:43:26 +0530274config FHANDLE
Andi Kleenf76be612016-04-01 14:31:40 -0700275 bool "open by fhandle syscalls" if EXPERT
Aneesh Kumar K.V990d6c22011-01-29 18:43:26 +0530276 select EXPORTFS
Andi Kleenf76be612016-04-01 14:31:40 -0700277 default y
Aneesh Kumar K.V990d6c22011-01-29 18:43:26 +0530278 help
279 If you say Y here, a user level program will be able to map
280 file names to handle and then later use the handle for
281 different file system operations. This is useful in implementing
282 userspace file servers, which now track files using handles instead
283 of names. The handle would remain the same even if file names
284 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
285 syscalls.
286
Josh Triplett69369a72014-04-03 14:48:27 -0700287config USELIB
288 bool "uselib syscall"
Riku Voipiob2113a42016-01-15 16:58:13 -0800289 def_bool ALPHA || M68K || SPARC || X86_32 || IA32_EMULATION
Josh Triplett69369a72014-04-03 14:48:27 -0700290 help
291 This option enables the uselib syscall, a system call used in the
292 dynamic linker from libc5 and earlier. glibc does not use this
293 system call. If you intend to run programs built on libc5 or
294 earlier, you may need to enable this syscall. Current systems
295 running glibc can safely disable this.
296
Linus Torvalds1da177e2005-04-16 15:20:36 -0700297config AUDIT
298 bool "Auditing support"
Chris Wright804a6a492005-05-11 10:52:45 +0100299 depends on NET
Linus Torvalds1da177e2005-04-16 15:20:36 -0700300 help
301 Enable auditing infrastructure that can be used with another
302 kernel subsystem, such as SELinux (which requires this for
Paul Moorecb74ed22016-01-13 09:18:55 -0500303 logging of avc messages output). System call auditing is included
304 on architectures which support it.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700305
AKASHI Takahiro7a017722014-02-25 18:16:24 +0900306config HAVE_ARCH_AUDITSYSCALL
307 bool
308
Linus Torvalds1da177e2005-04-16 15:20:36 -0700309config AUDITSYSCALL
Paul Moorecb74ed22016-01-13 09:18:55 -0500310 def_bool y
AKASHI Takahiro7a017722014-02-25 18:16:24 +0900311 depends on AUDIT && HAVE_ARCH_AUDITSYSCALL
Linus Torvalds1da177e2005-04-16 15:20:36 -0700312
Eric Paris939a67f2009-12-17 20:12:06 -0500313config AUDIT_WATCH
314 def_bool y
315 depends on AUDITSYSCALL
316 select FSNOTIFY
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
Eric Paris28a3a7e2009-12-17 20:12:05 -0500321 select FSNOTIFY
Al Viro74c3cbe2007-07-22 08:04:18 -0400322
Thomas Gleixnerd9817eb2010-09-27 12:45:59 +0000323source "kernel/irq/Kconfig"
Thomas Gleixner764e0da2012-05-21 23:16:18 +0200324source "kernel/time/Kconfig"
Thomas Gleixnerd9817eb2010-09-27 12:45:59 +0000325
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200326menu "CPU/Task time and stats accounting"
327
Frederic Weisbeckerabf917c2012-07-25 07:56:04 +0200328config VIRT_CPU_ACCOUNTING
329 bool
330
Frederic Weisbeckerfdf9c352012-09-09 14:56:31 +0200331choice
332 prompt "Cputime accounting"
333 default TICK_CPU_ACCOUNTING if !PPC64
Stephen Rothwell02fc8d32013-02-08 14:19:38 +1100334 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
Frederic Weisbeckerfdf9c352012-09-09 14:56:31 +0200335
336# Kind of a stub config for the pure tick based cputime accounting
337config TICK_CPU_ACCOUNTING
338 bool "Simple tick based cputime accounting"
Frederic Weisbeckerc58b0df2013-04-26 15:16:31 +0200339 depends on !S390 && !NO_HZ_FULL
Frederic Weisbeckerfdf9c352012-09-09 14:56:31 +0200340 help
341 This is the basic tick based cputime accounting that maintains
342 statistics about user, system and idle time spent on per jiffies
343 granularity.
344
345 If unsure, say Y.
346
Frederic Weisbeckerabf917c2012-07-25 07:56:04 +0200347config VIRT_CPU_ACCOUNTING_NATIVE
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200348 bool "Deterministic task and CPU time accounting"
Frederic Weisbeckerc58b0df2013-04-26 15:16:31 +0200349 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
Frederic Weisbeckerabf917c2012-07-25 07:56:04 +0200350 select VIRT_CPU_ACCOUNTING
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200351 help
352 Select this option to enable more accurate task and CPU time
353 accounting. This is done by reading a CPU counter on each
354 kernel entry and exit and on transitions within the kernel
355 between system, softirq and hardirq state, so there is a
356 small performance impact. In the case of s390 or IBM POWER > 5,
357 this also enables accounting of stolen time on logically-partitioned
358 systems.
359
Frederic Weisbeckerabf917c2012-07-25 07:56:04 +0200360config VIRT_CPU_ACCOUNTING_GEN
361 bool "Full dynticks CPU time accounting"
Kevin Hilmanff3fb252013-09-16 15:28:19 -0700362 depends on HAVE_CONTEXT_TRACKING
Kevin Hilman554b0002013-09-16 15:28:21 -0700363 depends on HAVE_VIRT_CPU_ACCOUNTING_GEN
Frederic Weisbeckerabf917c2012-07-25 07:56:04 +0200364 select VIRT_CPU_ACCOUNTING
365 select CONTEXT_TRACKING
366 help
367 Select this option to enable task and CPU time accounting on full
368 dynticks systems. This accounting is implemented by watching every
369 kernel-user boundaries using the context tracking subsystem.
370 The accounting is thus performed at the expense of some significant
371 overhead.
372
373 For now this is only useful if you are working on the full
374 dynticks subsystem development.
375
376 If unsure, say N.
377
Rik van Rielb58c3582016-07-13 16:50:02 +0200378endchoice
379
Frederic Weisbeckerfdf9c352012-09-09 14:56:31 +0200380config IRQ_TIME_ACCOUNTING
381 bool "Fine granularity task level IRQ time accounting"
Rik van Rielb58c3582016-07-13 16:50:02 +0200382 depends on HAVE_IRQ_TIME_ACCOUNTING && !VIRT_CPU_ACCOUNTING_NATIVE
Frederic Weisbeckerfdf9c352012-09-09 14:56:31 +0200383 help
384 Select this option to enable fine granularity task irq time
385 accounting. This is done by reading a timestamp on each
386 transitions between softirq and hardirq state, so there can be a
387 small performance impact.
388
389 If in doubt, say N here.
390
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200391config BSD_PROCESS_ACCT
392 bool "BSD Process Accounting"
Iulia Manda28138932015-04-15 16:16:41 -0700393 depends on MULTIUSER
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200394 help
395 If you say Y here, a user level program will be able to instruct the
396 kernel (via a special system call) to write process accounting
397 information to a file: whenever a process exits, information about
398 that process will be appended to the file by the kernel. The
399 information includes things such as creation time, owning user,
400 command name, memory usage, controlling terminal etc. (the complete
401 list is in the struct acct in <file:include/linux/acct.h>). It is
402 up to the user level program to do useful things with this
403 information. This is generally a good idea, so say Y.
404
405config BSD_PROCESS_ACCT_V3
406 bool "BSD Process Accounting version 3 file format"
407 depends on BSD_PROCESS_ACCT
408 default n
409 help
410 If you say Y here, the process accounting information is written
411 in a new file format that also logs the process IDs of each
412 process and it's parent. Note that this file format is incompatible
413 with previous v0/v1/v2 file formats, so you will need updated tools
414 for processing it. A preliminary version of these tools is available
415 at <http://www.gnu.org/software/acct/>.
416
417config TASKSTATS
Kees Cook19c92392012-10-02 11:19:29 -0700418 bool "Export task/process statistics through netlink"
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200419 depends on NET
Iulia Manda28138932015-04-15 16:16:41 -0700420 depends on MULTIUSER
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200421 default n
422 help
423 Export selected statistics for tasks/processes through the
424 generic netlink interface. Unlike BSD process accounting, the
425 statistics are available during the lifetime of tasks/processes as
426 responses to commands. Like BSD accounting, they are sent to user
427 space on task exit.
428
429 Say N if unsure.
430
431config TASK_DELAY_ACCT
Kees Cook19c92392012-10-02 11:19:29 -0700432 bool "Enable per-task delay accounting"
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200433 depends on TASKSTATS
Naveen N. Raof6db8342015-06-25 23:53:37 +0530434 select SCHED_INFO
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200435 help
436 Collect information on time spent by a task waiting for system
437 resources like cpu, synchronous block I/O completion and swapping
438 in pages. Such statistics can help in setting a task's priorities
439 relative to other tasks for cpu, io, rss limits etc.
440
441 Say N if unsure.
442
443config TASK_XACCT
Kees Cook19c92392012-10-02 11:19:29 -0700444 bool "Enable extended accounting over taskstats"
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200445 depends on TASKSTATS
446 help
447 Collect extended task accounting data and send the data
448 to userland for processing over the taskstats interface.
449
450 Say N if unsure.
451
452config TASK_IO_ACCOUNTING
Kees Cook19c92392012-10-02 11:19:29 -0700453 bool "Enable per-task storage I/O accounting"
Frederic Weisbecker391dc692012-09-09 14:22:07 +0200454 depends on TASK_XACCT
455 help
456 Collect information on the number of bytes of storage I/O which this
457 task has caused.
458
459 Say N if unsure.
460
461endmenu # "CPU/Task time and stats accounting"
462
Mike Travisc903ff82009-01-15 12:28:29 -0800463menu "RCU Subsystem"
464
Mike Travisc903ff82009-01-15 12:28:29 -0800465config TREE_RCU
Pranith Kumare72aeaf2015-04-21 17:29:42 -0400466 bool
467 default y if !PREEMPT && SMP
Mike Travisc903ff82009-01-15 12:28:29 -0800468 help
469 This option selects the RCU implementation that is
470 designed for very large SMP system with hundreds or
Paul E. McKenneyc17ef452009-06-23 17:12:47 -0700471 thousands of CPUs. It also scales down nicely to
472 smaller systems.
Mike Travisc903ff82009-01-15 12:28:29 -0800473
Pranith Kumar28f65692014-09-22 14:00:48 -0400474config PREEMPT_RCU
Pranith Kumare72aeaf2015-04-21 17:29:42 -0400475 bool
476 default y if PREEMPT
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700477 help
478 This option selects the RCU implementation that is
479 designed for very large SMP systems with hundreds or
480 thousands of CPUs, but for which real-time response
Paul E. McKenneybbe3eae2009-09-13 09:15:08 -0700481 is also required. It also scales down nicely to
482 smaller systems.
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700483
Paul E. McKenney9fc52d82013-01-08 15:48:33 -0800484 Select this option if you are unsure.
485
Paul E. McKenney9b1d82f2009-10-25 19:03:50 -0700486config TINY_RCU
Pranith Kumare72aeaf2015-04-21 17:29:42 -0400487 bool
488 default y if !PREEMPT && !SMP
Paul E. McKenney9b1d82f2009-10-25 19:03:50 -0700489 help
490 This option selects the RCU implementation that is
491 designed for UP systems from which real-time response
492 is not required. This option greatly reduces the
493 memory footprint of RCU.
494
Paul E. McKenney78cae102015-04-20 12:19:45 -0700495config RCU_EXPERT
496 bool "Make expert-level adjustments to RCU configuration"
497 default n
498 help
499 This option needs to be enabled if you wish to make
500 expert-level adjustments to RCU configuration. By default,
501 no such adjustments can be made, which has the often-beneficial
502 side-effect of preventing "make oldconfig" from asking you all
503 sorts of detailed questions about how you would like numerous
504 obscure RCU options to be set up.
505
506 Say Y if you need to make expert-level adjustments to RCU.
507
508 Say N if you are unsure.
509
Pranith Kumar83fe27e2014-12-05 11:24:45 -0500510config SRCU
511 bool
512 help
513 This option selects the sleepable version of RCU. This version
514 permits arbitrary sleeping or blocking within RCU read-side critical
515 sections.
516
Paul E. McKenney8315f422014-06-27 13:42:20 -0700517config TASKS_RCU
Paul E. McKenney82d0f4c2015-04-20 05:42:50 -0700518 bool
Paul E. McKenney8315f422014-06-27 13:42:20 -0700519 default n
Paul E. McKenney570dd3c2016-06-15 08:56:53 -0700520 depends on !UML
Pranith Kumar83fe27e2014-12-05 11:24:45 -0500521 select SRCU
Paul E. McKenney8315f422014-06-27 13:42:20 -0700522 help
523 This option enables a task-based RCU implementation that uses
524 only voluntary context switch (not preemption!), idle, and
525 user-mode execution as quiescent states.
526
Paul E. McKenney6bfc09e2012-10-19 12:49:17 -0700527config RCU_STALL_COMMON
Pranith Kumar28f65692014-09-22 14:00:48 -0400528 def_bool ( TREE_RCU || PREEMPT_RCU || RCU_TRACE )
Paul E. McKenney6bfc09e2012-10-19 12:49:17 -0700529 help
530 This option enables RCU CPU stall code that is common between
531 the TINY and TREE variants of RCU. The purpose is to allow
532 the tiny variants to disable RCU CPU stall warnings, while
533 making these warnings mandatory for the tree variants.
534
Frederic Weisbecker91d1aa432012-11-27 19:33:25 +0100535config CONTEXT_TRACKING
536 bool
537
Frederic Weisbecker91d1aa432012-11-27 19:33:25 +0100538config CONTEXT_TRACKING_FORCE
539 bool "Force context tracking"
540 depends on CONTEXT_TRACKING
Frederic Weisbeckerd84d27a2013-07-24 21:59:29 +0200541 default y if !NO_HZ_FULL
Frederic Weisbecker1fd2b442012-07-11 20:26:40 +0200542 help
Frederic Weisbeckerd84d27a2013-07-24 21:59:29 +0200543 The major pre-requirement for full dynticks to work is to
544 support the context tracking subsystem. But there are also
545 other dependencies to provide in order to make the full
546 dynticks working.
547
548 This option stands for testing when an arch implements the
549 context tracking backend but doesn't yet fullfill all the
550 requirements to make the full dynticks feature working.
551 Without the full dynticks, there is no way to test the support
552 for context tracking and the subsystems that rely on it: RCU
553 userspace extended quiescent state and tickless cputime
554 accounting. This option copes with the absence of the full
555 dynticks subsystem by forcing the context tracking on all
556 CPUs in the system.
557
Paul Gortmaker99c8b1e2013-10-24 10:07:47 -0400558 Say Y only if you're working on the development of an
Frederic Weisbeckerd84d27a2013-07-24 21:59:29 +0200559 architecture backend for the context tracking.
560
561 Say N otherwise, this option brings an overhead that you
562 don't want in production.
563
Frederic Weisbeckerd6771242012-10-11 01:48:28 +0200564
Mike Travisc903ff82009-01-15 12:28:29 -0800565config RCU_FANOUT
566 int "Tree-based hierarchical RCU fanout value"
567 range 2 64 if 64BIT
568 range 2 32 if !64BIT
Paul E. McKenney05c5df32015-04-20 14:27:43 -0700569 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
Mike Travisc903ff82009-01-15 12:28:29 -0800570 default 64 if 64BIT
571 default 32 if !64BIT
572 help
573 This option controls the fanout of hierarchical implementations
574 of RCU, allowing RCU to work efficiently on machines with
Paul E. McKenney4d87ffa2010-08-04 17:31:12 -0700575 large numbers of CPUs. This value must be at least the fourth
576 root of NR_CPUS, which allows NR_CPUS to be insanely large.
577 The default value of RCU_FANOUT should be used for production
578 systems, but if you are stress-testing the RCU implementation
579 itself, small RCU_FANOUT values allow you to test large-system
580 code paths on small(er) systems.
Mike Travisc903ff82009-01-15 12:28:29 -0800581
582 Select a specific number if testing RCU itself.
583 Take the default if unsure.
584
Paul E. McKenney8932a632012-04-19 12:20:14 -0700585config RCU_FANOUT_LEAF
586 int "Tree-based hierarchical RCU leaf-level fanout value"
Paul E. McKenney8739c5c2015-04-20 18:27:54 -0700587 range 2 64 if 64BIT
588 range 2 32 if !64BIT
Paul E. McKenney47d631a2015-04-21 09:12:13 -0700589 depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT
Paul E. McKenney8932a632012-04-19 12:20:14 -0700590 default 16
591 help
592 This option controls the leaf-level fanout of hierarchical
593 implementations of RCU, and allows trading off cache misses
594 against lock contention. Systems that synchronize their
595 scheduling-clock interrupts for energy-efficiency reasons will
596 want the default because the smaller leaf-level fanout keeps
597 lock contention levels acceptably low. Very large systems
598 (hundreds or thousands of CPUs) will instead want to set this
599 value to the maximum value possible in order to reduce the
600 number of cache misses incurred during RCU's grace-period
601 initialization. These systems tend to run CPU-bound, and thus
602 are not helped by synchronized interrupts, and thus tend to
603 skew them, which reduces lock contention enough that large
604 leaf-level fanouts work well.
605
606 Select a specific number if testing RCU itself.
607
608 Select the maximum permissible value for large systems.
609
610 Take the default if unsure.
611
Paul E. McKenney8bd93a22010-02-22 17:04:59 -0800612config RCU_FAST_NO_HZ
613 bool "Accelerate last non-dyntick-idle CPU's grace periods"
Paul E. McKenney78cae102015-04-20 12:19:45 -0700614 depends on NO_HZ_COMMON && SMP && RCU_EXPERT
Paul E. McKenney8bd93a22010-02-22 17:04:59 -0800615 default n
616 help
Paul E. McKenneyc0f4dfd2012-12-28 11:30:36 -0800617 This option permits CPUs to enter dynticks-idle state even if
618 they have RCU callbacks queued, and prevents RCU from waking
619 these CPUs up more than roughly once every four jiffies (by
620 default, you can adjust this using the rcutree.rcu_idle_gp_delay
621 parameter), thus improving energy efficiency. On the other
622 hand, this option increases the duration of RCU grace periods,
623 for example, slowing down synchronize_rcu().
Paul E. McKenney8bd93a22010-02-22 17:04:59 -0800624
Paul E. McKenneyc0f4dfd2012-12-28 11:30:36 -0800625 Say Y if energy efficiency is critically important, and you
626 don't care about increased grace-period durations.
Paul E. McKenney8bd93a22010-02-22 17:04:59 -0800627
628 Say N if you are unsure.
629
Mike Travisc903ff82009-01-15 12:28:29 -0800630config TREE_RCU_TRACE
Pranith Kumar28f65692014-09-22 14:00:48 -0400631 def_bool RCU_TRACE && ( TREE_RCU || PREEMPT_RCU )
Mike Travisc903ff82009-01-15 12:28:29 -0800632 select DEBUG_FS
633 help
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700634 This option provides tracing for the TREE_RCU and
Pranith Kumar28f65692014-09-22 14:00:48 -0400635 PREEMPT_RCU implementations, permitting Makefile to
Paul E. McKenneyf41d9112009-08-22 13:56:52 -0700636 trivially select kernel/rcutree_trace.c.
Mike Travisc903ff82009-01-15 12:28:29 -0800637
Paul E. McKenney24278d12010-09-27 17:25:23 -0700638config RCU_BOOST
639 bool "Enable RCU priority boosting"
Paul E. McKenney78cae102015-04-20 12:19:45 -0700640 depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT
Paul E. McKenney24278d12010-09-27 17:25:23 -0700641 default n
642 help
643 This option boosts the priority of preempted RCU readers that
644 block the current preemptible RCU grace period for too long.
645 This option also prevents heavy loads from blocking RCU
646 callback invocation for all flavors of RCU.
647
648 Say Y here if you are working with real-time apps or heavy loads
649 Say N here if you are unsure.
650
Clark Williams21871d72014-09-12 21:21:09 -0500651config RCU_KTHREAD_PRIO
652 int "Real-time priority to use for RCU worker threads"
Paul E. McKenneya94844b2014-12-12 07:37:48 -0800653 range 1 99 if RCU_BOOST
654 range 0 99 if !RCU_BOOST
655 default 1 if RCU_BOOST
656 default 0 if !RCU_BOOST
Paul E. McKenney26730f52015-04-21 09:22:14 -0700657 depends on RCU_EXPERT
Paul E. McKenney24278d12010-09-27 17:25:23 -0700658 help
Clark Williams21871d72014-09-12 21:21:09 -0500659 This option specifies the SCHED_FIFO priority value that will be
660 assigned to the rcuc/n and rcub/n threads and is also the value
661 used for RCU_BOOST (if enabled). If you are working with a
662 real-time application that has one or more CPU-bound threads
663 running at a real-time priority level, you should set
664 RCU_KTHREAD_PRIO to a priority higher than the highest-priority
665 real-time CPU-bound application thread. The default RCU_KTHREAD_PRIO
666 value of 1 is appropriate in the common case, which is real-time
Paul E. McKenneyc9336642012-04-18 16:20:18 -0700667 applications that do not have any CPU-bound threads.
668
669 Some real-time applications might not have a single real-time
670 thread that saturates a given CPU, but instead might have
671 multiple real-time threads that, taken together, fully utilize
Clark Williams21871d72014-09-12 21:21:09 -0500672 that CPU. In this case, you should set RCU_KTHREAD_PRIO to
Paul E. McKenneyc9336642012-04-18 16:20:18 -0700673 a priority higher than the lowest-priority thread that is
674 conspiring to prevent the CPU from running any non-real-time
675 tasks. For example, if one thread at priority 10 and another
676 thread at priority 5 are between themselves fully consuming
Clark Williams21871d72014-09-12 21:21:09 -0500677 the CPU time on a given CPU, then RCU_KTHREAD_PRIO should be
Paul E. McKenneyc9336642012-04-18 16:20:18 -0700678 set to priority 6 or higher.
Paul E. McKenney24278d12010-09-27 17:25:23 -0700679
680 Specify the real-time priority, or take the default if unsure.
681
682config RCU_BOOST_DELAY
683 int "Milliseconds to delay boosting after RCU grace-period start"
684 range 0 3000
685 depends on RCU_BOOST
686 default 500
687 help
688 This option specifies the time to wait after the beginning of
689 a given grace period before priority-boosting preempted RCU
690 readers blocking that grace period. Note that any RCU reader
691 blocking an expedited RCU grace period is boosted immediately.
692
693 Accept the default if unsure.
694
Paul E. McKenney3fbfbf72012-08-19 21:35:53 -0700695config RCU_NOCB_CPU
Paul E. McKenney9a5739d2013-03-28 20:48:36 -0700696 bool "Offload RCU callback processing from boot-selected CPUs"
Pranith Kumar28f65692014-09-22 14:00:48 -0400697 depends on TREE_RCU || PREEMPT_RCU
Paul E. McKenneybe55fa22015-06-02 05:29:18 -0700698 depends on RCU_EXPERT || NO_HZ_FULL
Paul E. McKenney3fbfbf72012-08-19 21:35:53 -0700699 default n
700 help
701 Use this option to reduce OS jitter for aggressive HPC or
702 real-time workloads. It can also be used to offload RCU
703 callback invocation to energy-efficient CPUs in battery-powered
704 asymmetric multiprocessors.
705
706 This option offloads callback invocation from the set of
707 CPUs specified at boot time by the rcu_nocbs parameter.
Paul E. McKenneya4889852012-12-03 08:16:28 -0800708 For each such CPU, a kthread ("rcuox/N") will be created to
709 invoke callbacks, where the "N" is the CPU being offloaded,
710 and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and
711 "s" for RCU-sched. Nothing prevents this kthread from running
712 on the specified CPUs, but (1) the kthreads may be preempted
713 between each callback, and (2) affinity or cgroups can be used
714 to force the kthreads to run on whatever set of CPUs is desired.
Paul E. McKenney3fbfbf72012-08-19 21:35:53 -0700715
Paul E. McKenney34ed62462013-01-07 13:37:42 -0800716 Say Y here if you want to help to debug reduced OS jitter.
Paul E. McKenney3fbfbf72012-08-19 21:35:53 -0700717 Say N here if you are unsure.
718
Paul E. McKenney911af502013-02-11 10:23:27 -0800719choice
720 prompt "Build-forced no-CBs CPUs"
721 default RCU_NOCB_CPU_NONE
Stefan Hengelein45687792014-09-02 19:55:11 +0200722 depends on RCU_NOCB_CPU
Paul E. McKenney911af502013-02-11 10:23:27 -0800723 help
Paul E. McKenney676c3dc2013-04-30 14:49:42 -0700724 This option allows no-CBs CPUs (whose RCU callbacks are invoked
725 from kthreads rather than from softirq context) to be specified
726 at build time. Additional no-CBs CPUs may be specified by
727 the rcu_nocbs= boot parameter.
Paul E. McKenney911af502013-02-11 10:23:27 -0800728
729config RCU_NOCB_CPU_NONE
730 bool "No build_forced no-CBs CPUs"
Paul E. McKenney911af502013-02-11 10:23:27 -0800731 help
732 This option does not force any of the CPUs to be no-CBs CPUs.
733 Only CPUs designated by the rcu_nocbs= boot parameter will be
Paul E. McKenney676c3dc2013-04-30 14:49:42 -0700734 no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU
735 kthreads whose names begin with "rcuo". All other CPUs will
736 invoke their own RCU callbacks in softirq context.
737
738 Select this option if you want to choose no-CBs CPUs at
739 boot time, for example, to allow testing of different no-CBs
740 configurations without having to rebuild the kernel each time.
Paul E. McKenney911af502013-02-11 10:23:27 -0800741
742config RCU_NOCB_CPU_ZERO
743 bool "CPU 0 is a build_forced no-CBs CPU"
Paul E. McKenney911af502013-02-11 10:23:27 -0800744 help
Paul E. McKenney676c3dc2013-04-30 14:49:42 -0700745 This option forces CPU 0 to be a no-CBs CPU, so that its RCU
746 callbacks are invoked by a per-CPU kthread whose name begins
747 with "rcuo". Additional CPUs may be designated as no-CBs
748 CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs.
749 All other CPUs will invoke their own RCU callbacks in softirq
750 context.
Paul E. McKenney911af502013-02-11 10:23:27 -0800751
752 Select this if CPU 0 needs to be a no-CBs CPU for real-time
Paul E. McKenney676c3dc2013-04-30 14:49:42 -0700753 or energy-efficiency reasons, but the real reason it exists
754 is to ensure that randconfig testing covers mixed systems.
Paul E. McKenney911af502013-02-11 10:23:27 -0800755
756config RCU_NOCB_CPU_ALL
757 bool "All CPUs are build_forced no-CBs CPUs"
Paul E. McKenney911af502013-02-11 10:23:27 -0800758 help
759 This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs=
Paul E. McKenney676c3dc2013-04-30 14:49:42 -0700760 boot parameter will be ignored. All CPUs' RCU callbacks will
761 be executed in the context of per-CPU rcuo kthreads created for
762 this purpose. Assuming that the kthreads whose names start with
763 "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter
764 on the remaining CPUs, but might decrease memory locality during
765 RCU-callback invocation, thus potentially degrading throughput.
Paul E. McKenney911af502013-02-11 10:23:27 -0800766
767 Select this if all CPUs need to be no-CBs CPUs for real-time
768 or energy-efficiency reasons.
769
770endchoice
771
Paul E. McKenneyee425712015-02-19 10:51:32 -0800772config RCU_EXPEDITE_BOOT
773 bool
774 default n
775 help
776 This option enables expedited grace periods at boot time,
777 as if rcu_expedite_gp() had been invoked early in boot.
778 The corresponding rcu_unexpedite_gp() is invoked from
779 rcu_end_inkernel_boot(), which is intended to be invoked
780 at the end of the kernel-only boot sequence, just before
781 init is exec'ed.
782
783 Accept the default if unsure.
784
Mike Travisc903ff82009-01-15 12:28:29 -0800785endmenu # "RCU Subsystem"
786
Vivek Goyalde5b56b2014-08-08 14:25:41 -0700787config BUILD_BIN2C
788 bool
789 default n
790
Linus Torvalds1da177e2005-04-16 15:20:36 -0700791config IKCONFIG
Ross Birof2443ab2006-09-30 23:27:25 -0700792 tristate "Kernel .config support"
Vivek Goyalde5b56b2014-08-08 14:25:41 -0700793 select BUILD_BIN2C
Linus Torvalds1da177e2005-04-16 15:20:36 -0700794 ---help---
795 This option enables the complete Linux kernel ".config" file
796 contents to be saved in the kernel. It provides documentation
797 of which kernel options are used in a running kernel or in an
798 on-disk kernel. This information can be extracted from the kernel
799 image file with the script scripts/extract-ikconfig and used as
800 input to rebuild the current kernel or to build another kernel.
801 It can also be extracted from a running kernel by reading
802 /proc/config.gz if enabled (below).
803
804config IKCONFIG_PROC
805 bool "Enable access to .config through /proc/config.gz"
806 depends on IKCONFIG && PROC_FS
807 ---help---
808 This option enables access to the kernel configuration file
809 through /proc/config.gz.
810
Alistair John Strachan794543a2007-05-08 00:31:15 -0700811config LOG_BUF_SHIFT
812 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
Ingo Molnarfb39f982015-07-01 10:19:11 +0200813 range 12 25
Adrian Bunkf17a32e2008-04-29 00:58:58 -0700814 default 17
Josh Triplett361e9df2014-10-03 16:00:54 -0700815 depends on PRINTK
Alistair John Strachan794543a2007-05-08 00:31:15 -0700816 help
Luis R. Rodriguez23b28992014-08-06 16:08:56 -0700817 Select the minimal kernel log buffer size as a power of 2.
818 The final size is affected by LOG_CPU_MAX_BUF_SHIFT config
819 parameter, see below. Any higher size also might be forced
820 by "log_buf_len" boot parameter.
821
Adrian Bunkf17a32e2008-04-29 00:58:58 -0700822 Examples:
Luis R. Rodriguez23b28992014-08-06 16:08:56 -0700823 17 => 128 KB
Adrian Bunkf17a32e2008-04-29 00:58:58 -0700824 16 => 64 KB
Luis R. Rodriguez23b28992014-08-06 16:08:56 -0700825 15 => 32 KB
826 14 => 16 KB
Alistair John Strachan794543a2007-05-08 00:31:15 -0700827 13 => 8 KB
828 12 => 4 KB
829
Luis R. Rodriguez23b28992014-08-06 16:08:56 -0700830config LOG_CPU_MAX_BUF_SHIFT
831 int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
Geert Uytterhoeven2240a312014-10-13 15:51:11 -0700832 depends on SMP
Luis R. Rodriguez23b28992014-08-06 16:08:56 -0700833 range 0 21
834 default 12 if !BASE_SMALL
835 default 0 if BASE_SMALL
Josh Triplett361e9df2014-10-03 16:00:54 -0700836 depends on PRINTK
Luis R. Rodriguez23b28992014-08-06 16:08:56 -0700837 help
838 This option allows to increase the default ring buffer size
839 according to the number of CPUs. The value defines the contribution
840 of each CPU as a power of 2. The used space is typically only few
841 lines however it might be much more when problems are reported,
842 e.g. backtraces.
843
844 The increased size means that a new buffer has to be allocated and
845 the original static one is unused. It makes sense only on systems
846 with more CPUs. Therefore this value is used only when the sum of
847 contributions is greater than the half of the default kernel ring
848 buffer as defined by LOG_BUF_SHIFT. The default values are set
849 so that more than 64 CPUs are needed to trigger the allocation.
850
851 Also this option is ignored when "log_buf_len" kernel parameter is
852 used as it forces an exact (power of two) size of the ring buffer.
853
854 The number of possible CPUs is used for this computation ignoring
Geert Uytterhoeven5e0d8d52016-06-05 10:47:02 +0200855 hotplugging making the computation optimal for the worst case
856 scenario while allowing a simple algorithm to be used from bootup.
Luis R. Rodriguez23b28992014-08-06 16:08:56 -0700857
858 Examples shift values and their meaning:
859 17 => 128 KB for each CPU
860 16 => 64 KB for each CPU
861 15 => 32 KB for each CPU
862 14 => 16 KB for each CPU
863 13 => 8 KB for each CPU
864 12 => 4 KB for each CPU
865
Petr Mladek427934b2016-05-20 17:00:39 -0700866config NMI_LOG_BUF_SHIFT
867 int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
868 range 10 21
869 default 13
870 depends on PRINTK_NMI
871 help
872 Select the size of a per-CPU buffer where NMI messages are temporary
873 stored. They are copied to the main log buffer in a safe context
874 to avoid a deadlock. The value defines the size as a power of 2.
875
876 NMI messages are rare and limited. The largest one is when
877 a backtrace is printed. It usually fits into 4KB. Select
878 8KB if you want to be on the safe side.
879
880 Examples:
881 17 => 128 KB for each CPU
882 16 => 64 KB for each CPU
883 15 => 32 KB for each CPU
884 14 => 16 KB for each CPU
885 13 => 8 KB for each CPU
886 12 => 4 KB for each CPU
887
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800888#
889# Architectures with an unreliable sched_clock() should select this:
890#
891config HAVE_UNSTABLE_SCHED_CLOCK
892 bool
893
Stephen Boyd38ff87f2013-06-01 23:39:40 -0700894config GENERIC_SCHED_CLOCK
895 bool
896
Andrea Arcangelibe3a7282012-10-04 01:50:47 +0200897#
898# For architectures that want to enable the support for NUMA-affine scheduler
899# balancing logic:
900#
901config ARCH_SUPPORTS_NUMA_BALANCING
902 bool
903
Peter Zijlstrabe5e6102013-11-18 18:27:06 +0100904#
Mel Gorman72b252a2015-09-04 15:47:32 -0700905# For architectures that prefer to flush all TLBs after a number of pages
906# are unmapped instead of sending one IPI per page to flush. The architecture
907# must provide guarantees on what happens if a clean TLB cache entry is
908# written after the unmap. Details are in mm/rmap.c near the check for
909# should_defer_flush. The architecture should also consider if the full flush
910# and the refill costs are offset by the savings of sending fewer IPIs.
911config ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
912 bool
913
914#
Peter Zijlstrabe5e6102013-11-18 18:27:06 +0100915# For architectures that know their GCC __int128 support is sound
916#
917config ARCH_SUPPORTS_INT128
918 bool
919
Andrea Arcangelibe3a7282012-10-04 01:50:47 +0200920# For architectures that (ab)use NUMA to represent different memory regions
921# all cpu-local but of different latencies, such as SuperH.
922#
923config ARCH_WANT_NUMA_VARIABLE_LOCALITY
924 bool
925
Andrea Arcangelibe3a7282012-10-04 01:50:47 +0200926config NUMA_BALANCING
927 bool "Memory placement aware NUMA scheduler"
Andrea Arcangelibe3a7282012-10-04 01:50:47 +0200928 depends on ARCH_SUPPORTS_NUMA_BALANCING
929 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
930 depends on SMP && NUMA && MIGRATION
931 help
932 This option adds support for automatic NUMA aware memory/task placement.
933 The mechanism is quite primitive and is based on migrating memory when
Paul Gortmaker6d56a412013-08-13 11:06:50 -0400934 it has references to the node the task is running on.
Andrea Arcangelibe3a7282012-10-04 01:50:47 +0200935
936 This system will be inactive on UMA systems.
937
Aneesh Kumar K.V6f7c97e2014-12-10 15:43:37 -0800938config NUMA_BALANCING_DEFAULT_ENABLED
939 bool "Automatically enable NUMA aware memory/task placement"
940 default y
941 depends on NUMA_BALANCING
942 help
943 If set, automatic NUMA balancing will be enabled if running on a NUMA
944 machine.
945
Li Zefan23964d22009-01-15 13:50:58 -0800946menuconfig CGROUPS
Christoph Jaeger6341e622014-12-20 15:41:11 -0500947 bool "Control Group support"
Tejun Heo2bd59d42014-02-11 11:52:49 -0500948 select KERNFS
Paul Menageddbcc7e2007-10-18 23:39:30 -0700949 help
Li Zefan23964d22009-01-15 13:50:58 -0800950 This option adds support for grouping sets of processes together, for
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800951 use with process control subsystems such as Cpusets, CFS, memory
952 controls or device isolation.
953 See
KAMEZAWA Hiroyuki5cdc38f2009-01-07 18:07:30 -0800954 - Documentation/scheduler/sched-design-CFS.txt (CFS)
seokhoon.yoon9991a9c2016-08-02 14:03:13 -0700955 - Documentation/cgroup-v1/ (features for grouping, isolation
Li Zefan45ce80f2009-01-15 13:50:59 -0800956 and resource control)
Paul Menageddbcc7e2007-10-18 23:39:30 -0700957
958 Say N if unsure.
959
Li Zefan23964d22009-01-15 13:50:58 -0800960if CGROUPS
961
Johannes Weiner3e32cb22014-12-10 15:42:31 -0800962config PAGE_COUNTER
963 bool
964
Andrew Mortonc255a452012-07-31 16:43:02 -0700965config MEMCG
Johannes Weinera0166ec2015-12-17 17:19:56 -0500966 bool "Memory controller"
Johannes Weiner3e32cb22014-12-10 15:42:31 -0800967 select PAGE_COUNTER
Tejun Heo79bd9812013-11-22 18:20:42 -0500968 select EVENTFD
Balbir Singh00f0b822008-03-04 14:28:39 -0800969 help
Johannes Weinera0166ec2015-12-17 17:19:56 -0500970 Provides control over the memory footprint of tasks in a cgroup.
Balbir Singh00f0b822008-03-04 14:28:39 -0800971
Andrew Mortonc255a452012-07-31 16:43:02 -0700972config MEMCG_SWAP
Johannes Weinera0166ec2015-12-17 17:19:56 -0500973 bool "Swap controller"
Andrew Mortonc255a452012-07-31 16:43:02 -0700974 depends on MEMCG && SWAP
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -0800975 help
Johannes Weinera0166ec2015-12-17 17:19:56 -0500976 Provides control over the swap space consumed by tasks in a cgroup.
977
Andrew Mortonc255a452012-07-31 16:43:02 -0700978config MEMCG_SWAP_ENABLED
Johannes Weinera0166ec2015-12-17 17:19:56 -0500979 bool "Swap controller enabled by default"
Andrew Mortonc255a452012-07-31 16:43:02 -0700980 depends on MEMCG_SWAP
Michal Hockoa42c3902010-11-24 12:57:08 -0800981 default y
982 help
983 Memory Resource Controller Swap Extension comes with its price in
984 a bigger memory consumption. General purpose distribution kernels
Jim Cromie43d547f2010-12-17 14:32:36 -0700985 which want to enable the feature but keep it disabled by default
Michal Hocko07555ac2013-08-22 16:35:46 -0700986 and let the user enable it by swapaccount=1 boot command line
Michal Hockoa42c3902010-11-24 12:57:08 -0800987 parameter should have this option unselected.
988 For those who want to have the feature enabled by default should
989 select this option (if, for some reason, they need to disable it
WANG Cong00a66d22011-07-25 17:12:12 -0700990 then swapaccount=0 does the trick).
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -0800991
Johannes Weiner6bf024e2015-12-17 17:19:57 -0500992config BLK_CGROUP
993 bool "IO controller"
994 depends on BLOCK
Aneesh Kumar K.V2bc64a22012-07-31 16:42:12 -0700995 default n
Johannes Weiner6bf024e2015-12-17 17:19:57 -0500996 ---help---
997 Generic block IO controller cgroup interface. This is the common
998 cgroup interface which should be used by various IO controlling
999 policies.
Aneesh Kumar K.V2bc64a22012-07-31 16:42:12 -07001000
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001001 Currently, CFQ IO scheduler uses it to recognize task groups and
1002 control disk bandwidth allocation (proportional time slice allocation)
1003 to such task groups. It is also used by bio throttling logic in
1004 block layer to implement upper limit in IO rates on a device.
Stephane Eraniane5d13672011-02-14 11:20:01 +02001005
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001006 This option only enables generic Block IO controller infrastructure.
1007 One needs to also enable actual IO controlling logic/policy. For
1008 enabling proportional weight division of disk bandwidth in CFQ, set
1009 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
1010 CONFIG_BLK_DEV_THROTTLING=y.
1011
seokhoon.yoon9991a9c2016-08-02 14:03:13 -07001012 See Documentation/cgroup-v1/blkio-controller.txt for more information.
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001013
1014config DEBUG_BLK_CGROUP
1015 bool "IO controller debugging"
1016 depends on BLK_CGROUP
1017 default n
1018 ---help---
1019 Enable some debugging help. Currently it exports additional stat
1020 files in a cgroup which can be useful for debugging.
1021
1022config CGROUP_WRITEBACK
1023 bool
1024 depends on MEMCG && BLK_CGROUP
1025 default y
Stephane Eraniane5d13672011-02-14 11:20:01 +02001026
Dhaval Giani7c941432010-01-20 13:26:18 +01001027menuconfig CGROUP_SCHED
Johannes Weinera0166ec2015-12-17 17:19:56 -05001028 bool "CPU controller"
Dhaval Giani7c941432010-01-20 13:26:18 +01001029 default n
1030 help
1031 This feature lets CPU scheduler recognize task groups and control CPU
1032 bandwidth allocation to such task groups. It uses cgroups to group
1033 tasks.
1034
1035if CGROUP_SCHED
1036config FAIR_GROUP_SCHED
1037 bool "Group scheduling for SCHED_OTHER"
1038 depends on CGROUP_SCHED
1039 default CGROUP_SCHED
1040
Paul Turnerab84d312011-07-21 09:43:28 -07001041config CFS_BANDWIDTH
1042 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
Paul Turnerab84d312011-07-21 09:43:28 -07001043 depends on FAIR_GROUP_SCHED
1044 default n
1045 help
1046 This option allows users to define CPU bandwidth rates (limits) for
1047 tasks running within the fair group scheduler. Groups with no limit
1048 set are considered to be unconstrained and will run with no
1049 restriction.
1050 See tip/Documentation/scheduler/sched-bwc.txt for more information.
1051
Dhaval Giani7c941432010-01-20 13:26:18 +01001052config RT_GROUP_SCHED
1053 bool "Group scheduling for SCHED_RR/FIFO"
Dhaval Giani7c941432010-01-20 13:26:18 +01001054 depends on CGROUP_SCHED
1055 default n
1056 help
1057 This feature lets you explicitly allocate real CPU bandwidth
Li Zefan32bd7eb2010-03-24 13:17:19 +08001058 to task groups. If enabled, it will also make it impossible to
Dhaval Giani7c941432010-01-20 13:26:18 +01001059 schedule realtime tasks for non-root users until you allocate
1060 realtime bandwidth for them.
1061 See Documentation/scheduler/sched-rt-group.txt for more information.
1062
1063endif #CGROUP_SCHED
1064
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001065config CGROUP_PIDS
1066 bool "PIDs controller"
1067 help
1068 Provides enforcement of process number limits in the scope of a
1069 cgroup. Any attempt to fork more processes than is allowed in the
1070 cgroup will fail. PIDs are fundamentally a global resource because it
1071 is fairly trivial to reach PID exhaustion before you reach even a
1072 conservative kmemcg limit. As a result, it is possible to grind a
1073 system to halt without being limited by other cgroup policies. The
Parav Pandit6cc578d2016-03-05 11:30:56 +05301074 PIDs controller is designed to stop this from happening.
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001075
1076 It should be noted that organisational operations (such as attaching
Parav Pandit6cc578d2016-03-05 11:30:56 +05301077 to a cgroup hierarchy will *not* be blocked by the PIDs controller),
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001078 since the PIDs limit only affects a process's ability to fork, not to
1079 attach to a cgroup.
1080
1081config CGROUP_FREEZER
1082 bool "Freezer controller"
1083 help
1084 Provides a way to freeze and unfreeze all tasks in a
1085 cgroup.
1086
Johannes Weiner489c2a22016-01-20 15:02:41 -08001087 This option affects the ORIGINAL cgroup interface. The cgroup2 memory
1088 controller includes important in-kernel memory consumers per default.
1089
1090 If you're using cgroup2, say N.
1091
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001092config CGROUP_HUGETLB
1093 bool "HugeTLB controller"
1094 depends on HUGETLB_PAGE
1095 select PAGE_COUNTER
Vivek Goyalafc24d42010-04-26 19:27:56 +02001096 default n
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001097 help
1098 Provides a cgroup controller for HugeTLB pages.
1099 When you enable this, you can put a per cgroup limit on HugeTLB usage.
1100 The limit is enforced during page fault. Since HugeTLB doesn't
1101 support page reclaim, enforcing the limit at page fault time implies
1102 that, the application will get SIGBUS signal if it tries to access
1103 HugeTLB pages beyond its limit. This requires the application to know
1104 beforehand how much HugeTLB pages it would require for its use. The
1105 control group is tracked in the third page lru pointer. This means
1106 that we cannot use the controller with huge page less than 3 pages.
Vivek Goyalafc24d42010-04-26 19:27:56 +02001107
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001108config CPUSETS
1109 bool "Cpuset controller"
1110 help
1111 This option will let you create and manage CPUSETs which
1112 allow dynamically partitioning a system into sets of CPUs and
1113 Memory Nodes and assigning tasks to run only within those sets.
1114 This is primarily useful on large SMP or NUMA systems.
Vivek Goyalafc24d42010-04-26 19:27:56 +02001115
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001116 Say N if unsure.
Vivek Goyalafc24d42010-04-26 19:27:56 +02001117
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001118config PROC_PID_CPUSET
1119 bool "Include legacy /proc/<pid>/cpuset file"
1120 depends on CPUSETS
Tejun Heo89e9b9e2015-05-22 17:13:36 -04001121 default y
1122
Johannes Weiner6bf024e2015-12-17 17:19:57 -05001123config CGROUP_DEVICE
1124 bool "Device controller"
1125 help
1126 Provides a cgroup controller implementing whitelists for
1127 devices which a process in the cgroup can mknod or open.
1128
1129config CGROUP_CPUACCT
1130 bool "Simple CPU accounting controller"
1131 help
1132 Provides a simple controller for monitoring the
1133 total CPU consumed by the tasks in a cgroup.
1134
1135config CGROUP_PERF
1136 bool "Perf controller"
1137 depends on PERF_EVENTS
1138 help
1139 This option extends the perf per-cpu mode to restrict monitoring
1140 to threads which belong to the cgroup specified and run on the
1141 designated cpu.
1142
1143 Say N if unsure.
1144
1145config CGROUP_DEBUG
1146 bool "Example controller"
1147 default n
1148 help
1149 This option enables a simple controller that exports
1150 debugging information about the cgroups framework.
1151
1152 Say N.
1153
Li Zefan23964d22009-01-15 13:50:58 -08001154endif # CGROUPS
KAMEZAWA Hiroyukic0777192009-01-07 18:07:57 -08001155
Cyrill Gorcunov067bce12012-01-12 17:20:49 -08001156config CHECKPOINT_RESTORE
1157 bool "Checkpoint/restore support" if EXPERT
Iago López Galeiras2e13ba52015-06-25 15:00:57 -07001158 select PROC_CHILDREN
Cyrill Gorcunov067bce12012-01-12 17:20:49 -08001159 default n
1160 help
1161 Enables additional kernel features in a sake of checkpoint/restore.
1162 In particular it adds auxiliary prctl codes to setup process text,
1163 data and heap segment sizes, and a few additional /proc filesystem
1164 entries.
1165
1166 If unsure, say N here.
1167
Daniel Lezcano8dd2a822010-10-27 15:34:38 -07001168menuconfig NAMESPACES
David Rientjes6a108a12011-01-20 14:44:16 -08001169 bool "Namespaces support" if EXPERT
Iulia Manda28138932015-04-15 16:16:41 -07001170 depends on MULTIUSER
David Rientjes6a108a12011-01-20 14:44:16 -08001171 default !EXPERT
Pavel Emelyanovc5289a692008-02-08 04:18:19 -08001172 help
1173 Provides the way to make tasks work with different objects using
1174 the same id. For example same IPC id may refer to different objects
1175 or same user id or pid may refer to different tasks when used in
1176 different namespaces.
1177
Daniel Lezcano8dd2a822010-10-27 15:34:38 -07001178if NAMESPACES
1179
Pavel Emelyanov58bfdd6d2008-02-08 04:18:21 -08001180config UTS_NS
1181 bool "UTS namespace"
Daniel Lezcano17a6d442010-10-27 15:34:37 -07001182 default y
Pavel Emelyanov58bfdd6d2008-02-08 04:18:21 -08001183 help
1184 In this namespace tasks see different info provided with the
1185 uname() system call
1186
Pavel Emelyanovae5e1b22008-02-08 04:18:22 -08001187config IPC_NS
1188 bool "IPC namespace"
Daniel Lezcano8dd2a822010-10-27 15:34:38 -07001189 depends on (SYSVIPC || POSIX_MQUEUE)
Daniel Lezcano17a6d442010-10-27 15:34:37 -07001190 default y
Pavel Emelyanovae5e1b22008-02-08 04:18:22 -08001191 help
1192 In this namespace tasks work with IPC ids which correspond to
Serge E. Hallyn614b84c2009-04-06 19:01:08 -07001193 different IPC objects in different namespaces.
Pavel Emelyanovae5e1b22008-02-08 04:18:22 -08001194
Pavel Emelyanovaee16ce2008-02-08 04:18:23 -08001195config USER_NS
Kees Cook19c92392012-10-02 11:19:29 -07001196 bool "User namespace"
Eric W. Biederman5673a942011-11-17 10:23:55 -08001197 default n
Pavel Emelyanovaee16ce2008-02-08 04:18:23 -08001198 help
1199 This allows containers, i.e. vservers, to use user namespaces
1200 to provide different user info for different servers.
Eric W. Biedermane11f0ae2013-01-25 16:48:31 -08001201
1202 When user namespaces are enabled in the kernel it is
Johannes Weinerd886f4e2016-01-20 15:02:47 -08001203 recommended that the MEMCG option also be enabled and that
1204 user-space use the memory control groups to limit the amount
1205 of memory a memory unprivileged users can use.
Eric W. Biedermane11f0ae2013-01-25 16:48:31 -08001206
Pavel Emelyanovaee16ce2008-02-08 04:18:23 -08001207 If unsure, say N.
1208
Pavel Emelyanov74bd59b2008-02-08 04:18:24 -08001209config PID_NS
Daniel Lezcano9bd38c22010-10-27 15:34:37 -07001210 bool "PID Namespaces"
Daniel Lezcano17a6d442010-10-27 15:34:37 -07001211 default y
Pavel Emelyanov74bd59b2008-02-08 04:18:24 -08001212 help
Heikki Orsila12d2b8f2008-07-06 15:48:02 +03001213 Support process id namespaces. This allows having multiple
Matt LaPlante692105b2009-01-26 11:12:25 +01001214 processes with the same pid as long as they are in different
Pavel Emelyanov74bd59b2008-02-08 04:18:24 -08001215 pid namespaces. This is a building block of containers.
1216
Matt Helsleyd6eb6332009-01-26 12:25:55 -08001217config NET_NS
1218 bool "Network namespace"
Daniel Lezcano8dd2a822010-10-27 15:34:38 -07001219 depends on NET
Daniel Lezcano17a6d442010-10-27 15:34:37 -07001220 default y
Matt Helsleyd6eb6332009-01-26 12:25:55 -08001221 help
1222 Allow user space to create what appear to be multiple instances
1223 of the network stack.
1224
Daniel Lezcano8dd2a822010-10-27 15:34:38 -07001225endif # NAMESPACES
1226
Mike Galbraith5091faa2010-11-30 14:18:03 +01001227config SCHED_AUTOGROUP
1228 bool "Automatic process group scheduling"
Mike Galbraith5091faa2010-11-30 14:18:03 +01001229 select CGROUPS
1230 select CGROUP_SCHED
1231 select FAIR_GROUP_SCHED
1232 help
1233 This option optimizes the scheduler for common desktop workloads by
1234 automatically creating and populating task groups. This separation
1235 of workloads isolates aggressive CPU burners (like build jobs) from
1236 desktop applications. Task group autogeneration is currently based
1237 upon task session.
1238
Daniel Lezcano7af37be2010-10-27 15:34:41 -07001239config SYSFS_DEPRECATED
Ferenc Wagner5d6a4ea2011-01-10 19:04:22 +01001240 bool "Enable deprecated sysfs features to support old userspace tools"
Daniel Lezcano7af37be2010-10-27 15:34:41 -07001241 depends on SYSFS
1242 default n
1243 help
1244 This option adds code that switches the layout of the "block" class
1245 devices, to not show up in /sys/class/block/, but only in
1246 /sys/block/.
1247
1248 This switch is only active when the sysfs.deprecated=1 boot option is
1249 passed or the SYSFS_DEPRECATED_V2 option is set.
1250
1251 This option allows new kernels to run on old distributions and tools,
1252 which might get confused by /sys/class/block/. Since 2007/2008 all
1253 major distributions and tools handle this just fine.
1254
1255 Recent distributions and userspace tools after 2009/2010 depend on
1256 the existence of /sys/class/block/, and will not work with this
1257 option enabled.
1258
1259 Only if you are using a new kernel on an old distribution, you might
1260 need to say Y here.
1261
1262config SYSFS_DEPRECATED_V2
Ferenc Wagner5d6a4ea2011-01-10 19:04:22 +01001263 bool "Enable deprecated sysfs features by default"
Daniel Lezcano7af37be2010-10-27 15:34:41 -07001264 default n
1265 depends on SYSFS
1266 depends on SYSFS_DEPRECATED
1267 help
1268 Enable deprecated sysfs by default.
1269
1270 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1271 option.
1272
1273 Only if you are using a new kernel on an old distribution, you might
1274 need to say Y here. Even then, odds are you would not need it
1275 enabled, you can always pass the boot option if absolutely necessary.
1276
1277config RELAY
1278 bool "Kernel->user space relay support (formerly relayfs)"
1279 help
1280 This option enables support for relay interface support in
1281 certain file systems (such as debugfs).
1282 It is designed to provide an efficient mechanism for tools and
1283 facilities to relay large amounts of data from kernel space to
1284 user space.
1285
1286 If unsure, say N.
1287
Dimitri Gorokhovikf9916332007-03-06 01:42:17 -08001288config BLK_DEV_INITRD
1289 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1290 depends on BROKEN || !FRV
1291 help
1292 The initial RAM filesystem is a ramfs which is loaded by the
1293 boot loader (loadlin or lilo) and that is mounted as root
1294 before the normal boot procedure. It is typically used to
1295 load modules needed to mount the "real" root file system,
1296 etc. See <file:Documentation/initrd.txt> for details.
1297
1298 If RAM disk support (BLK_DEV_RAM) is also included, this
1299 also enables initial RAM disk (initrd) support and adds
1300 15 Kbytes (more on some other architectures) to the kernel size.
1301
1302 If unsure say Y.
1303
Jean-Paul Samanc33df4e2007-02-10 01:44:43 -08001304if BLK_DEV_INITRD
1305
Sam Ravnborgdbec4862005-08-10 20:44:50 +02001306source "usr/Kconfig"
1307
Jean-Paul Samanc33df4e2007-02-10 01:44:43 -08001308endif
1309
Arnd Bergmann877417e2016-04-25 17:35:27 +02001310choice
1311 prompt "Compiler optimization level"
1312 default CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE
1313
1314config CC_OPTIMIZE_FOR_PERFORMANCE
1315 bool "Optimize for performance"
1316 help
1317 This is the default optimization level for the kernel, building
1318 with the "-O2" compiler flag for best performance and most
1319 helpful compile-time warnings.
1320
Linus Torvaldsc45b4f12005-12-14 18:52:21 -08001321config CC_OPTIMIZE_FOR_SIZE
Ingo Molnar96fffeb2008-04-28 01:39:43 +02001322 bool "Optimize for size"
Linus Torvaldsc45b4f12005-12-14 18:52:21 -08001323 help
Masahiro Yamada31a4af72014-08-05 14:43:07 +09001324 Enabling this option will pass "-Os" instead of "-O2" to
1325 your compiler resulting in a smaller kernel.
Linus Torvaldsc45b4f12005-12-14 18:52:21 -08001326
Kirill Smelkov3a55fb02012-11-02 15:41:01 +04001327 If unsure, say N.
Linus Torvaldsc45b4f12005-12-14 18:52:21 -08001328
Arnd Bergmann877417e2016-04-25 17:35:27 +02001329endchoice
1330
Randy Dunlap08470622006-09-30 23:28:13 -07001331config SYSCTL
1332 bool
1333
Randy Dunlapb943c462009-03-10 12:55:46 -07001334config ANON_INODES
1335 bool
1336
Mike Frysinger657a5202013-04-30 15:28:45 -07001337config HAVE_UID16
1338 bool
1339
1340config SYSCTL_EXCEPTION_TRACE
1341 bool
1342 help
1343 Enable support for /proc/sys/debug/exception-trace.
1344
1345config SYSCTL_ARCH_UNALIGN_NO_WARN
1346 bool
1347 help
1348 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1349 Allows arch to define/use @no_unaligned_warning to possibly warn
1350 about unaligned access emulation going on under the hood.
1351
1352config SYSCTL_ARCH_UNALIGN_ALLOW
1353 bool
1354 help
1355 Enable support for /proc/sys/kernel/unaligned-trap
1356 Allows arches to define/use @unaligned_enabled to runtime toggle
1357 the unaligned access emulation.
1358 see arch/parisc/kernel/unaligned.c for reference
1359
Mike Frysinger657a5202013-04-30 15:28:45 -07001360config HAVE_PCSPKR_PLATFORM
1361 bool
1362
Alexei Starovoitovf89b7752014-10-23 18:41:08 -07001363# interpreter that classic socket filters depend on
1364config BPF
1365 bool
1366
David Rientjes6a108a12011-01-20 14:44:16 -08001367menuconfig EXPERT
1368 bool "Configure standard kernel features (expert users)"
Josh Triplettf505c552011-06-05 18:23:58 -07001369 # Unhide debug options, to make the on-by-default options visible
1370 select DEBUG_KERNEL
Linus Torvalds1da177e2005-04-16 15:20:36 -07001371 help
1372 This option allows certain base kernel options and settings
1373 to be disabled or tweaked. This is for specialized
1374 environments which can tolerate a "non-standard" kernel.
1375 Only use this if you really know what you are doing.
1376
Chuck Ebbertae81f9e2006-09-16 12:15:53 -07001377config UID16
David Rientjes6a108a12011-01-20 14:44:16 -08001378 bool "Enable 16-bit UID system calls" if EXPERT
Iulia Manda28138932015-04-15 16:16:41 -07001379 depends on HAVE_UID16 && MULTIUSER
Chuck Ebbertae81f9e2006-09-16 12:15:53 -07001380 default y
1381 help
1382 This enables the legacy 16-bit UID syscall wrappers.
1383
Iulia Manda28138932015-04-15 16:16:41 -07001384config MULTIUSER
1385 bool "Multiple users, groups and capabilities support" if EXPERT
1386 default y
1387 help
1388 This option enables support for non-root users, groups and
1389 capabilities.
1390
1391 If you say N here, all processes will run with UID 0, GID 0, and all
1392 possible capabilities. Saying N here also compiles out support for
1393 system calls related to UIDs, GIDs, and capabilities, such as setuid,
1394 setgid, and capset.
1395
1396 If unsure, say Y here.
1397
Fabian Frederickf6187762014-06-04 16:11:12 -07001398config SGETMASK_SYSCALL
1399 bool "sgetmask/ssetmask syscalls support" if EXPERT
1400 def_bool PARISC || MN10300 || BLACKFIN || M68K || PPC || MIPS || X86 || SPARC || CRIS || MICROBLAZE || SUPERH
1401 ---help---
1402 sys_sgetmask and sys_ssetmask are obsolete system calls
1403 no longer supported in libc but still enabled by default in some
1404 architectures.
1405
1406 If unsure, leave the default option here.
1407
Fabian Frederick6af9f7b2014-04-03 14:48:25 -07001408config SYSFS_SYSCALL
1409 bool "Sysfs syscall support" if EXPERT
1410 default y
1411 ---help---
1412 sys_sysfs is an obsolete system call no longer supported in libc.
1413 Note that disabling this option is more secure but might break
1414 compatibility with some systems.
1415
1416 If unsure say Y here.
1417
Eric W. Biedermanb89a8172006-09-27 01:51:04 -07001418config SYSCTL_SYSCALL
David Rientjes6a108a12011-01-20 14:44:16 -08001419 bool "Sysctl syscall support" if EXPERT
Eric W. Biederman26a70342009-11-05 05:26:41 -08001420 depends on PROC_SYSCTL
WANG Congc736de62011-11-02 13:39:25 -07001421 default n
Eric W. Biedermanb89a8172006-09-27 01:51:04 -07001422 select SYSCTL
1423 ---help---
Eric W. Biederman13bb7e32006-11-08 17:44:51 -08001424 sys_sysctl uses binary paths that have been found challenging
1425 to properly maintain and use. The interface in /proc/sys
1426 using paths with ascii names is now the primary path to this
1427 information.
Eric W. Biedermanb89a8172006-09-27 01:51:04 -07001428
Eric W. Biederman13bb7e32006-11-08 17:44:51 -08001429 Almost nothing using the binary sysctl interface so if you are
1430 trying to save some space it is probably safe to disable this,
1431 making your kernel marginally smaller.
Eric W. Biedermanb89a8172006-09-27 01:51:04 -07001432
WANG Congc736de62011-11-02 13:39:25 -07001433 If unsure say N here.
Chuck Ebbertae81f9e2006-09-16 12:15:53 -07001434
Linus Torvalds1da177e2005-04-16 15:20:36 -07001435config KALLSYMS
David Rientjes6a108a12011-01-20 14:44:16 -08001436 bool "Load all symbols for debugging/ksymoops" if EXPERT
Linus Torvalds1da177e2005-04-16 15:20:36 -07001437 default y
1438 help
1439 Say Y here to let the kernel print out symbolic crash information and
1440 symbolic stack backtraces. This increases the size of the kernel
1441 somewhat, as all symbols have to be loaded into the kernel image.
1442
1443config KALLSYMS_ALL
1444 bool "Include all symbols in kallsyms"
1445 depends on DEBUG_KERNEL && KALLSYMS
1446 help
Artem Bityutskiy71a83ec2011-04-05 13:24:57 +03001447 Normally kallsyms only contains the symbols of functions for nicer
1448 OOPS messages and backtraces (i.e., symbols from the text and inittext
1449 sections). This is sufficient for most cases. And only in very rare
1450 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1451 names of variables from the data sections, etc).
Linus Torvalds1da177e2005-04-16 15:20:36 -07001452
Artem Bityutskiy71a83ec2011-04-05 13:24:57 +03001453 This option makes sure that all symbols are loaded into the kernel
1454 image (i.e., symbols from all sections) in cost of increased kernel
1455 size (depending on the kernel configuration, it may be 300KiB or
1456 something like this).
Linus Torvalds1da177e2005-04-16 15:20:36 -07001457
Artem Bityutskiy71a83ec2011-04-05 13:24:57 +03001458 Say N unless you really need all symbols.
Matt Mackalld59745c2005-05-01 08:59:02 -07001459
Ard Biesheuvel4d5d5662016-03-15 14:58:12 -07001460config KALLSYMS_ABSOLUTE_PERCPU
1461 bool
Randy Dunlap076501f2016-07-06 16:06:53 -07001462 depends on KALLSYMS
Ard Biesheuvel4d5d5662016-03-15 14:58:12 -07001463 default X86_64 && SMP
1464
Ard Biesheuvel2213e9a2016-03-15 14:58:19 -07001465config KALLSYMS_BASE_RELATIVE
1466 bool
1467 depends on KALLSYMS
1468 default !IA64 && !(TILE && 64BIT)
1469 help
1470 Instead of emitting them as absolute values in the native word size,
1471 emit the symbol references in the kallsyms table as 32-bit entries,
1472 each containing a relative value in the range [base, base + U32_MAX]
1473 or, when KALLSYMS_ABSOLUTE_PERCPU is in effect, each containing either
1474 an absolute value in the range [0, S32_MAX] or a relative value in the
1475 range [base, base + S32_MAX], where base is the lowest relative symbol
1476 address encountered in the image.
1477
1478 On 64-bit builds, this reduces the size of the address table by 50%,
1479 but more importantly, it results in entries whose values are build
1480 time constants, and no relocation pass is required at runtime to fix
1481 up the entries based on the runtime load address of the kernel.
1482
Matt Mackalld59745c2005-05-01 08:59:02 -07001483config PRINTK
1484 default y
David Rientjes6a108a12011-01-20 14:44:16 -08001485 bool "Enable support for printk" if EXPERT
Frederic Weisbecker74876a92012-10-12 18:00:23 +02001486 select IRQ_WORK
Matt Mackalld59745c2005-05-01 08:59:02 -07001487 help
1488 This option enables normal printk support. Removing it
1489 eliminates most of the message strings from the kernel image
1490 and makes the kernel more or less silent. As this makes it
1491 very difficult to diagnose system problems, saying N here is
1492 strongly discouraged.
1493
Petr Mladek42a0bb32016-05-20 17:00:33 -07001494config PRINTK_NMI
1495 def_bool y
1496 depends on PRINTK
1497 depends on HAVE_NMI
1498
Matt Mackallc8538a72005-05-01 08:59:01 -07001499config BUG
David Rientjes6a108a12011-01-20 14:44:16 -08001500 bool "BUG() support" if EXPERT
Matt Mackallc8538a72005-05-01 08:59:01 -07001501 default y
1502 help
1503 Disabling this option eliminates support for BUG and WARN, reducing
1504 the size of your kernel image and potentially quietly ignoring
1505 numerous fatal conditions. You should only consider disabling this
1506 option for embedded systems with no facilities for reporting errors.
1507 Just say Y.
1508
Matt Mackall708e9a72006-01-08 01:05:25 -08001509config ELF_CORE
Alex Kelly046d6622012-10-04 17:15:23 -07001510 depends on COREDUMP
Matt Mackall708e9a72006-01-08 01:05:25 -08001511 default y
David Rientjes6a108a12011-01-20 14:44:16 -08001512 bool "Enable ELF core dumps" if EXPERT
Matt Mackall708e9a72006-01-08 01:05:25 -08001513 help
1514 Enable support for generating core dumps. Disabling saves about 4k.
1515
Ralf Baechle8761f1a2011-06-01 19:05:09 +01001516
Stas Sergeeve5e1d3c2008-05-07 12:39:56 +02001517config PCSPKR_PLATFORM
David Rientjes6a108a12011-01-20 14:44:16 -08001518 bool "Enable PC-Speaker support" if EXPERT
Ralf Baechle8761f1a2011-06-01 19:05:09 +01001519 depends on HAVE_PCSPKR_PLATFORM
Ralf Baechle15f304b2011-06-01 19:04:59 +01001520 select I8253_LOCK
Stas Sergeeve5e1d3c2008-05-07 12:39:56 +02001521 default y
1522 help
1523 This option allows to disable the internal PC-Speaker
1524 support, saving some memory.
1525
Linus Torvalds1da177e2005-04-16 15:20:36 -07001526config BASE_FULL
1527 default y
David Rientjes6a108a12011-01-20 14:44:16 -08001528 bool "Enable full-sized data structures for core" if EXPERT
Linus Torvalds1da177e2005-04-16 15:20:36 -07001529 help
1530 Disabling this option reduces the size of miscellaneous core
1531 kernel data structures. This saves memory on small machines,
1532 but may reduce performance.
1533
1534config FUTEX
David Rientjes6a108a12011-01-20 14:44:16 -08001535 bool "Enable futex support" if EXPERT
Linus Torvalds1da177e2005-04-16 15:20:36 -07001536 default y
Ingo Molnar23f78d4a2006-06-27 02:54:53 -07001537 select RT_MUTEXES
Linus Torvalds1da177e2005-04-16 15:20:36 -07001538 help
1539 Disabling this option will cause the kernel to be built without
1540 support for "fast userspace mutexes". The resulting kernel may not
1541 run glibc-based applications correctly.
1542
Heiko Carstens03b8c7b2014-03-02 13:09:47 +01001543config HAVE_FUTEX_CMPXCHG
1544 bool
Josh Triplett62b4d202014-10-03 16:19:24 -07001545 depends on FUTEX
Heiko Carstens03b8c7b2014-03-02 13:09:47 +01001546 help
1547 Architectures should select this if futex_atomic_cmpxchg_inatomic()
1548 is implemented and always working. This removes a couple of runtime
1549 checks.
1550
Linus Torvalds1da177e2005-04-16 15:20:36 -07001551config EPOLL
David Rientjes6a108a12011-01-20 14:44:16 -08001552 bool "Enable eventpoll support" if EXPERT
Linus Torvalds1da177e2005-04-16 15:20:36 -07001553 default y
Adrian Bunk448e3ce2007-07-31 00:39:10 -07001554 select ANON_INODES
Linus Torvalds1da177e2005-04-16 15:20:36 -07001555 help
1556 Disabling this option will cause the kernel to be built without
1557 support for epoll family of system calls.
1558
Davide Libenzifba2afa2007-05-10 22:23:13 -07001559config SIGNALFD
David Rientjes6a108a12011-01-20 14:44:16 -08001560 bool "Enable signalfd() system call" if EXPERT
Adrian Bunk448e3ce2007-07-31 00:39:10 -07001561 select ANON_INODES
Davide Libenzifba2afa2007-05-10 22:23:13 -07001562 default y
1563 help
1564 Enable the signalfd() system call that allows to receive signals
1565 on a file descriptor.
1566
1567 If unsure, say Y.
1568
Davide Libenzib215e282007-05-10 22:23:16 -07001569config TIMERFD
David Rientjes6a108a12011-01-20 14:44:16 -08001570 bool "Enable timerfd() system call" if EXPERT
Adrian Bunk448e3ce2007-07-31 00:39:10 -07001571 select ANON_INODES
Davide Libenzib215e282007-05-10 22:23:16 -07001572 default y
1573 help
1574 Enable the timerfd() system call that allows to receive timer
1575 events on a file descriptor.
1576
1577 If unsure, say Y.
1578
Davide Libenzie1ad7462007-05-10 22:23:19 -07001579config EVENTFD
David Rientjes6a108a12011-01-20 14:44:16 -08001580 bool "Enable eventfd() system call" if EXPERT
Adrian Bunk448e3ce2007-07-31 00:39:10 -07001581 select ANON_INODES
Davide Libenzie1ad7462007-05-10 22:23:19 -07001582 default y
1583 help
1584 Enable the eventfd() system call that allows to receive both
1585 kernel notification (ie. KAIO) or userspace notifications.
1586
1587 If unsure, say Y.
1588
Alexei Starovoitovf89b7752014-10-23 18:41:08 -07001589# syscall, maps, verifier
1590config BPF_SYSCALL
Ingo Molnare1abf2c2015-04-02 15:51:39 +02001591 bool "Enable bpf() system call"
Alexei Starovoitovf89b7752014-10-23 18:41:08 -07001592 select ANON_INODES
1593 select BPF
1594 default n
1595 help
1596 Enable the bpf() system call that allows to manipulate eBPF
1597 programs and maps via file descriptors.
1598
Linus Torvalds1da177e2005-04-16 15:20:36 -07001599config SHMEM
David Rientjes6a108a12011-01-20 14:44:16 -08001600 bool "Use full shmem filesystem" if EXPERT
Linus Torvalds1da177e2005-04-16 15:20:36 -07001601 default y
1602 depends on MMU
1603 help
1604 The shmem is an internal filesystem used to manage shared memory.
1605 It is backed by swap and manages resource limits. It is also exported
1606 to userspace as tmpfs if TMPFS is enabled. Disabling this
1607 option replaces shmem and tmpfs with the much simpler ramfs code,
1608 which may be appropriate on small systems without swap.
1609
Thomas Petazzoniebf3f092008-10-15 22:05:12 -07001610config AIO
David Rientjes6a108a12011-01-20 14:44:16 -08001611 bool "Enable AIO support" if EXPERT
Thomas Petazzoniebf3f092008-10-15 22:05:12 -07001612 default y
1613 help
1614 This option enables POSIX asynchronous I/O which may by used
Mike Frysinger657a5202013-04-30 15:28:45 -07001615 by some high performance threaded applications. Disabling
1616 this option saves about 7k.
1617
Josh Triplettd3ac21c2014-08-17 19:41:09 -05001618config ADVISE_SYSCALLS
1619 bool "Enable madvise/fadvise syscalls" if EXPERT
1620 default y
1621 help
1622 This option enables the madvise and fadvise syscalls, used by
1623 applications to advise the kernel about their future memory or file
1624 usage, improving performance. If building an embedded system where no
1625 applications use these syscalls, you can disable this option to save
1626 space.
1627
Andrea Arcangelia14c1512015-09-04 15:46:54 -07001628config USERFAULTFD
1629 bool "Enable userfaultfd() system call"
1630 select ANON_INODES
Andrea Arcangelia14c1512015-09-04 15:46:54 -07001631 depends on MMU
1632 help
1633 Enable the userfaultfd() system call that allows to intercept and
1634 handle page faults in userland.
1635
Mike Frysinger657a5202013-04-30 15:28:45 -07001636config PCI_QUIRKS
1637 default y
1638 bool "Enable PCI quirk workarounds" if EXPERT
1639 depends on PCI
1640 help
1641 This enables workarounds for various PCI chipset
1642 bugs/quirks. Disable this only if your target machine is
1643 unaffected by PCI quirks.
Thomas Petazzoniebf3f092008-10-15 22:05:12 -07001644
Mathieu Desnoyers5b25b132015-09-11 13:07:39 -07001645config MEMBARRIER
1646 bool "Enable membarrier() system call" if EXPERT
1647 default y
1648 help
1649 Enable the membarrier() system call that allows issuing memory
1650 barriers across all running threads, which can be used to distribute
1651 the cost of user-space memory barriers asymmetrically by transforming
1652 pairs of memory barriers into pairs consisting of membarrier() and a
1653 compiler barrier.
1654
1655 If unsure, say Y.
1656
Randy Dunlap6befe5f2011-04-26 12:33:21 -07001657config EMBEDDED
1658 bool "Embedded system"
Josh Triplett5d2acfc2014-04-07 15:39:09 -07001659 option allnoconfig_y
Randy Dunlap6befe5f2011-04-26 12:33:21 -07001660 select EXPERT
1661 help
1662 This option should be enabled if compiling the kernel for
1663 an embedded system so certain expert options are available
1664 for configuration.
1665
Ingo Molnarcdd6c482009-09-21 12:02:48 +02001666config HAVE_PERF_EVENTS
Thomas Gleixner0793a612008-12-04 20:12:29 +01001667 bool
Mike Frysinger018df722009-06-12 13:17:43 -04001668 help
1669 See tools/perf/design.txt for details.
Thomas Gleixner0793a612008-12-04 20:12:29 +01001670
Peter Zijlstra906010b2009-09-21 16:08:49 +02001671config PERF_USE_VMALLOC
1672 bool
1673 help
1674 See tools/perf/design.txt for details
1675
Ingo Molnar57c0c152009-09-21 12:20:38 +02001676menu "Kernel Performance Events And Counters"
Thomas Gleixner0793a612008-12-04 20:12:29 +01001677
Ingo Molnarcdd6c482009-09-21 12:02:48 +02001678config PERF_EVENTS
Ingo Molnar57c0c152009-09-21 12:20:38 +02001679 bool "Kernel performance events and counters"
Robert Richter392d65a2012-04-05 18:24:44 +02001680 default y if PROFILING
Ingo Molnarcdd6c482009-09-21 12:02:48 +02001681 depends on HAVE_PERF_EVENTS
Ingo Molnar4c59e462008-12-08 19:38:33 +01001682 select ANON_INODES
Peter Zijlstrae360adb2010-10-14 14:01:34 +08001683 select IRQ_WORK
Pranith Kumar83fe27e2014-12-05 11:24:45 -05001684 select SRCU
Thomas Gleixner0793a612008-12-04 20:12:29 +01001685 help
Ingo Molnar57c0c152009-09-21 12:20:38 +02001686 Enable kernel support for various performance events provided
1687 by software and hardware.
Thomas Gleixner0793a612008-12-04 20:12:29 +01001688
Thadeu Lima de Souza Cascardodd770382009-10-30 19:32:25 -02001689 Software events are supported either built-in or via the
Ingo Molnar57c0c152009-09-21 12:20:38 +02001690 use of generic tracepoints.
1691
1692 Most modern CPUs support performance events via performance
1693 counter registers. These registers count the number of certain
Thomas Gleixner0793a612008-12-04 20:12:29 +01001694 types of hw events: such as instructions executed, cachemisses
1695 suffered, or branches mis-predicted - without slowing down the
1696 kernel or applications. These registers can also trigger interrupts
1697 when a threshold number of events have passed - and can thus be
1698 used to profile the code that runs on that CPU.
1699
Ingo Molnar57c0c152009-09-21 12:20:38 +02001700 The Linux Performance Event subsystem provides an abstraction of
Thadeu Lima de Souza Cascardodd770382009-10-30 19:32:25 -02001701 these software and hardware event capabilities, available via a
Ingo Molnar57c0c152009-09-21 12:20:38 +02001702 system call and used by the "perf" utility in tools/perf/. It
Thomas Gleixner0793a612008-12-04 20:12:29 +01001703 provides per task and per CPU counters, and it provides event
1704 capabilities on top of those.
1705
1706 Say Y if unsure.
1707
Peter Zijlstra906010b2009-09-21 16:08:49 +02001708config DEBUG_PERF_USE_VMALLOC
1709 default n
1710 bool "Debug: use vmalloc to back perf mmap() buffers"
Michael Ellermancb3071132015-05-04 16:26:39 +10001711 depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
Peter Zijlstra906010b2009-09-21 16:08:49 +02001712 select PERF_USE_VMALLOC
1713 help
1714 Use vmalloc memory to back perf mmap() buffers.
1715
1716 Mostly useful for debugging the vmalloc code on platforms
1717 that don't require it.
1718
1719 Say N if unsure.
1720
Thomas Gleixner0793a612008-12-04 20:12:29 +01001721endmenu
1722
Christoph Lameterf8891e52006-06-30 01:55:45 -07001723config VM_EVENT_COUNTERS
1724 default y
David Rientjes6a108a12011-01-20 14:44:16 -08001725 bool "Enable VM event counters for /proc/vmstat" if EXPERT
Christoph Lameterf8891e52006-06-30 01:55:45 -07001726 help
Paul Jackson2aea4fb2006-12-22 01:06:10 -08001727 VM event counters are needed for event counts to be shown.
1728 This option allows the disabling of the VM event counters
David Rientjes6a108a12011-01-20 14:44:16 -08001729 on EXPERT systems. /proc/vmstat will only show page counts
Paul Jackson2aea4fb2006-12-22 01:06:10 -08001730 if VM event counters are disabled.
Christoph Lameterf8891e52006-06-30 01:55:45 -07001731
Christoph Lameter41ecc552007-05-09 02:32:44 -07001732config SLUB_DEBUG
1733 default y
David Rientjes6a108a12011-01-20 14:44:16 -08001734 bool "Enable SLUB debugging support" if EXPERT
Christoph Lameterf6acb632008-04-29 16:16:06 -07001735 depends on SLUB && SYSFS
Christoph Lameter41ecc552007-05-09 02:32:44 -07001736 help
1737 SLUB has extensive debug support features. Disabling these can
1738 result in significant savings in code size. This also disables
1739 SLUB sysfs support. /sys/slab will not exist and there will be
1740 no support for cache validation etc.
1741
Randy Dunlapb943c462009-03-10 12:55:46 -07001742config COMPAT_BRK
1743 bool "Disable heap randomization"
1744 default y
1745 help
1746 Randomizing heap placement makes heap exploits harder, but it
1747 also breaks ancient binaries (including anything libc5 based).
1748 This option changes the bootup default to heap randomization
Matt LaPlante692105b2009-01-26 11:12:25 +01001749 disabled, and can be overridden at runtime by setting
Randy Dunlapb943c462009-03-10 12:55:46 -07001750 /proc/sys/kernel/randomize_va_space to 2.
1751
1752 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1753
Christoph Lameter81819f02007-05-06 14:49:36 -07001754choice
1755 prompt "Choose SLAB allocator"
Christoph Lametera0acd822007-07-17 04:03:32 -07001756 default SLUB
Christoph Lameter81819f02007-05-06 14:49:36 -07001757 help
1758 This option allows to select a slab allocator.
1759
1760config SLAB
1761 bool "SLAB"
1762 help
1763 The regular slab allocator that is established and known to work
Christoph Lameter34013882007-05-09 02:32:47 -07001764 well in all environments. It organizes cache hot objects in
Simon Arlott02f56212008-11-05 22:18:19 +00001765 per cpu and per node queues.
Christoph Lameter81819f02007-05-06 14:49:36 -07001766
1767config SLUB
Christoph Lameter81819f02007-05-06 14:49:36 -07001768 bool "SLUB (Unqueued Allocator)"
1769 help
1770 SLUB is a slab allocator that minimizes cache line usage
1771 instead of managing queues of cached objects (SLAB approach).
1772 Per cpu caching is realized using slabs of objects instead
1773 of queues of objects. SLUB can use memory efficiently
Simon Arlott02f56212008-11-05 22:18:19 +00001774 and has enhanced diagnostics. SLUB is the default choice for
1775 a slab allocator.
Christoph Lameter81819f02007-05-06 14:49:36 -07001776
1777config SLOB
David Rientjes6a108a12011-01-20 14:44:16 -08001778 depends on EXPERT
Christoph Lameter81819f02007-05-06 14:49:36 -07001779 bool "SLOB (Simple Allocator)"
1780 help
Matt Mackall37291452008-02-04 22:29:38 -08001781 SLOB replaces the stock allocator with a drastically simpler
1782 allocator. SLOB is generally more space efficient but
1783 does not perform as well on large systems.
Christoph Lameter81819f02007-05-06 14:49:36 -07001784
1785endchoice
1786
Thomas Garnierc7ce4f602016-05-19 17:10:37 -07001787config SLAB_FREELIST_RANDOM
1788 default n
Thomas Garnier210e7a42016-07-26 15:21:59 -07001789 depends on SLAB || SLUB
Thomas Garnierc7ce4f602016-05-19 17:10:37 -07001790 bool "SLAB freelist randomization"
1791 help
Thomas Garnier210e7a42016-07-26 15:21:59 -07001792 Randomizes the freelist order used on creating new pages. This
Thomas Garnierc7ce4f602016-05-19 17:10:37 -07001793 security feature reduces the predictability of the kernel slab
1794 allocator against heap overflows.
1795
Joonsoo Kim345c9052013-06-19 14:05:52 +09001796config SLUB_CPU_PARTIAL
1797 default y
Uwe Kleine-Königb39ffbf2013-07-17 16:54:59 +02001798 depends on SLUB && SMP
Joonsoo Kim345c9052013-06-19 14:05:52 +09001799 bool "SLUB per cpu partial cache"
1800 help
1801 Per cpu partial caches accellerate objects allocation and freeing
1802 that is local to a processor at the price of more indeterminism
1803 in the latency of the free. On overflow these caches will be cleared
1804 which requires the taking of locks that may cause latency spikes.
1805 Typically one would choose no for a realtime system.
1806
Jie Zhangea637632009-12-14 18:00:02 -08001807config MMAP_ALLOW_UNINITIALIZED
1808 bool "Allow mmapped anonymous memory to be uninitialized"
David Rientjes6a108a12011-01-20 14:44:16 -08001809 depends on EXPERT && !MMU
Jie Zhangea637632009-12-14 18:00:02 -08001810 default n
1811 help
1812 Normally, and according to the Linux spec, anonymous memory obtained
1813 from mmap() has it's contents cleared before it is passed to
1814 userspace. Enabling this config option allows you to request that
1815 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1816 providing a huge performance boost. If this option is not enabled,
1817 then the flag will be ignored.
1818
1819 This is taken advantage of by uClibc's malloc(), and also by
1820 ELF-FDPIC binfmt's brk and stack allocator.
1821
1822 Because of the obvious security issues, this option should only be
1823 enabled on embedded devices where you control what is run in
1824 userspace. Since that isn't generally a problem on no-MMU systems,
1825 it is normally safe to say Y here.
1826
1827 See Documentation/nommu-mmap.txt for more information.
1828
David Howells091f6e22015-07-20 21:16:28 +01001829config SYSTEM_DATA_VERIFICATION
1830 def_bool n
1831 select SYSTEM_TRUSTED_KEYRING
1832 select KEYS
1833 select CRYPTO
David Howellsd43de6c2016-03-03 21:49:27 +00001834 select CRYPTO_RSA
David Howells091f6e22015-07-20 21:16:28 +01001835 select ASYMMETRIC_KEY_TYPE
1836 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
David Howells091f6e22015-07-20 21:16:28 +01001837 select ASN1
1838 select OID_REGISTRY
1839 select X509_CERTIFICATE_PARSER
1840 select PKCS7_MESSAGE_PARSER
Peter Foley82c04ff2014-04-18 15:07:11 -07001841 help
David Howells091f6e22015-07-20 21:16:28 +01001842 Provide PKCS#7 message verification using the contents of the system
1843 trusted keyring to provide public keys. This then can be used for
1844 module verification, kexec image verification and firmware blob
1845 verification.
Peter Foley82c04ff2014-04-18 15:07:11 -07001846
Mathieu Desnoyers125e5642008-02-02 15:10:36 -05001847config PROFILING
Robert Richterb309a292010-02-26 15:01:23 +01001848 bool "Profiling support"
Mathieu Desnoyers125e5642008-02-02 15:10:36 -05001849 help
1850 Say Y here to enable the extended profiling support mechanisms used
1851 by profilers such as OProfile.
1852
Ingo Molnar5f87f112008-07-23 14:15:22 +02001853#
1854# Place an empty function call at each tracepoint site. Can be
1855# dynamically changed for a probe function.
1856#
Mathieu Desnoyers97e1c182008-07-18 12:16:16 -04001857config TRACEPOINTS
Ingo Molnar5f87f112008-07-23 14:15:22 +02001858 bool
Mathieu Desnoyers97e1c182008-07-18 12:16:16 -04001859
Mathieu Desnoyersfb32e032008-02-02 15:10:33 -05001860source "arch/Kconfig"
1861
Linus Torvalds1da177e2005-04-16 15:20:36 -07001862endmenu # General setup
1863
Dmitry Baryshkovee7e5512008-06-29 14:18:46 +04001864config HAVE_GENERIC_DMA_COHERENT
1865 bool
1866 default n
1867
Linus Torvalds158a9622008-01-02 13:04:48 -08001868config SLABINFO
1869 bool
1870 depends on PROC_FS
Christoph Lameter0f389ec2008-04-14 18:53:02 +03001871 depends on SLAB || SLUB_DEBUG
Linus Torvalds158a9622008-01-02 13:04:48 -08001872 default y
1873
Chuck Ebbertae81f9e2006-09-16 12:15:53 -07001874config RT_MUTEXES
Christoph Jaeger6341e622014-12-20 15:41:11 -05001875 bool
Chuck Ebbertae81f9e2006-09-16 12:15:53 -07001876
Linus Torvalds1da177e2005-04-16 15:20:36 -07001877config BASE_SMALL
1878 int
1879 default 0 if BASE_FULL
1880 default 1 if !BASE_FULL
1881
Jan Engelhardt66da5732007-07-15 23:39:29 -07001882menuconfig MODULES
Linus Torvalds1da177e2005-04-16 15:20:36 -07001883 bool "Enable loadable module support"
Yann E. MORIN11097a02013-08-11 16:07:50 +02001884 option modules
Linus Torvalds1da177e2005-04-16 15:20:36 -07001885 help
1886 Kernel modules are small pieces of compiled code which can
1887 be inserted in the running kernel, rather than being
1888 permanently built into the kernel. You use the "modprobe"
1889 tool to add (and sometimes remove) them. If you say Y here,
1890 many parts of the kernel can be built as modules (by
1891 answering M instead of Y where indicated): this is most
1892 useful for infrequently used options which are not required
1893 for booting. For more information, see the man pages for
1894 modprobe, lsmod, modinfo, insmod and rmmod.
1895
1896 If you say Y here, you will need to run "make
1897 modules_install" to put the modules under /lib/modules/
1898 where modprobe can find them (you may need to be root to do
1899 this).
1900
1901 If unsure, say Y.
1902
Robert P. J. Day0b0de142008-08-04 13:31:32 -04001903if MODULES
1904
Linus Torvalds826e4502008-05-04 17:04:16 -07001905config MODULE_FORCE_LOAD
1906 bool "Forced module loading"
Linus Torvalds826e4502008-05-04 17:04:16 -07001907 default n
1908 help
Rusty Russell91e37a72008-05-09 16:25:28 +10001909 Allow loading of modules without version information (ie. modprobe
1910 --force). Forced module loading sets the 'F' (forced) taint flag and
1911 is usually a really bad idea.
Linus Torvalds826e4502008-05-04 17:04:16 -07001912
Linus Torvalds1da177e2005-04-16 15:20:36 -07001913config MODULE_UNLOAD
1914 bool "Module unloading"
Linus Torvalds1da177e2005-04-16 15:20:36 -07001915 help
1916 Without this option you will not be able to unload any
1917 modules (note that some modules may not be unloadable
Denys Vlasenkof7f5b672008-07-22 19:24:26 -05001918 anyway), which makes your kernel smaller, faster
1919 and simpler. If unsure, say Y.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001920
1921config MODULE_FORCE_UNLOAD
1922 bool "Forced module unloading"
Kees Cook19c92392012-10-02 11:19:29 -07001923 depends on MODULE_UNLOAD
Linus Torvalds1da177e2005-04-16 15:20:36 -07001924 help
1925 This option allows you to force a module to unload, even if the
1926 kernel believes it is unsafe: the kernel will remove the module
1927 without waiting for anyone to stop using it (using the -f option to
1928 rmmod). This is mainly for kernel developers and desperate users.
1929 If unsure, say N.
1930
Linus Torvalds1da177e2005-04-16 15:20:36 -07001931config MODVERSIONS
Sam Ravnborg0d541642005-12-26 23:04:02 +01001932 bool "Module versioning support"
Linus Torvalds1da177e2005-04-16 15:20:36 -07001933 help
1934 Usually, you have to use modules compiled with your kernel.
1935 Saying Y here makes it sometimes possible to use modules
1936 compiled for different kernels, by adding enough information
1937 to the modules to (hopefully) spot any changes which would
1938 make them incompatible with the kernel you are running. If
1939 unsure, say N.
1940
1941config MODULE_SRCVERSION_ALL
1942 bool "Source checksum for all modules"
Linus Torvalds1da177e2005-04-16 15:20:36 -07001943 help
1944 Modules which contain a MODULE_VERSION get an extra "srcversion"
1945 field inserted into their modinfo section, which contains a
1946 sum of the source files which made it. This helps maintainers
1947 see exactly which source was used to build a module (since
1948 others sometimes change the module source without updating
1949 the version). With this option, such a "srcversion" field
1950 will be created for all modules. If unsure, say N.
1951
Rusty Russell106a4ee2012-09-26 10:09:40 +01001952config MODULE_SIG
1953 bool "Module signature verification"
1954 depends on MODULES
David Howells091f6e22015-07-20 21:16:28 +01001955 select SYSTEM_DATA_VERIFICATION
Rusty Russell106a4ee2012-09-26 10:09:40 +01001956 help
1957 Check modules for valid signatures upon load: the signature
1958 is simply appended to the module. For more information see
1959 Documentation/module-signing.txt.
1960
David Howells228c37f2015-08-11 12:38:54 +01001961 Note that this option adds the OpenSSL development packages as a
1962 kernel build dependency so that the signing tool can use its crypto
1963 library.
1964
David Howellsea0b6dc2012-09-26 10:09:50 +01001965 !!!WARNING!!! If you enable this option, you MUST make sure that the
1966 module DOES NOT get stripped after being signed. This includes the
1967 debuginfo strip done by some packagers (such as rpmbuild) and
1968 inclusion into an initramfs that wants the module size reduced.
1969
Rusty Russell106a4ee2012-09-26 10:09:40 +01001970config MODULE_SIG_FORCE
1971 bool "Require modules to be validly signed"
1972 depends on MODULE_SIG
1973 help
1974 Reject unsigned modules or signed modules for which we don't have a
1975 key. Without this, such modules will simply taint the kernel.
David Howellsea0b6dc2012-09-26 10:09:50 +01001976
Michal Marekd9d8d7e2013-01-25 13:41:31 +10301977config MODULE_SIG_ALL
1978 bool "Automatically sign all modules"
1979 default y
1980 depends on MODULE_SIG
1981 help
1982 Sign all modules during make modules_install. Without this option,
1983 modules must be signed manually, using the scripts/sign-file tool.
1984
1985comment "Do not forget to sign required modules with scripts/sign-file"
1986 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
1987
David Howellsea0b6dc2012-09-26 10:09:50 +01001988choice
1989 prompt "Which hash algorithm should modules be signed with?"
1990 depends on MODULE_SIG
1991 help
1992 This determines which sort of hashing algorithm will be used during
1993 signature generation. This algorithm _must_ be built into the kernel
1994 directly so that signature verification can take place. It is not
1995 possible to load a signed module containing the algorithm to check
1996 the signature on that module.
1997
1998config MODULE_SIG_SHA1
1999 bool "Sign modules with SHA-1"
2000 select CRYPTO_SHA1
2001
2002config MODULE_SIG_SHA224
2003 bool "Sign modules with SHA-224"
2004 select CRYPTO_SHA256
2005
2006config MODULE_SIG_SHA256
2007 bool "Sign modules with SHA-256"
2008 select CRYPTO_SHA256
2009
2010config MODULE_SIG_SHA384
2011 bool "Sign modules with SHA-384"
2012 select CRYPTO_SHA512
2013
2014config MODULE_SIG_SHA512
2015 bool "Sign modules with SHA-512"
2016 select CRYPTO_SHA512
2017
2018endchoice
2019
Michal Marek22753672013-01-25 13:41:00 +10302020config MODULE_SIG_HASH
2021 string
2022 depends on MODULE_SIG
2023 default "sha1" if MODULE_SIG_SHA1
2024 default "sha224" if MODULE_SIG_SHA224
2025 default "sha256" if MODULE_SIG_SHA256
2026 default "sha384" if MODULE_SIG_SHA384
2027 default "sha512" if MODULE_SIG_SHA512
2028
Bertrand Jacquinbeb50df2014-08-27 20:31:56 +09302029config MODULE_COMPRESS
2030 bool "Compress modules on installation"
2031 depends on MODULES
2032 help
Bertrand Jacquinbeb50df2014-08-27 20:31:56 +09302033
Rusty Russellb6c09b52015-06-16 12:16:22 +09302034 Compresses kernel modules when 'make modules_install' is run; gzip or
2035 xz depending on "Compression algorithm" below.
Bertrand Jacquinbeb50df2014-08-27 20:31:56 +09302036
Rusty Russellb6c09b52015-06-16 12:16:22 +09302037 module-init-tools MAY support gzip, and kmod MAY support gzip and xz.
Bertrand Jacquinbeb50df2014-08-27 20:31:56 +09302038
Rusty Russellb6c09b52015-06-16 12:16:22 +09302039 Out-of-tree kernel modules installed using Kbuild will also be
2040 compressed upon installation.
Bertrand Jacquinbeb50df2014-08-27 20:31:56 +09302041
Rusty Russellb6c09b52015-06-16 12:16:22 +09302042 Note: for modules inside an initrd or initramfs, it's more efficient
2043 to compress the whole initrd or initramfs instead.
Bertrand Jacquinbeb50df2014-08-27 20:31:56 +09302044
Rusty Russellb6c09b52015-06-16 12:16:22 +09302045 Note: This is fully compatible with signed modules.
2046
2047 If in doubt, say N.
Bertrand Jacquinbeb50df2014-08-27 20:31:56 +09302048
2049choice
2050 prompt "Compression algorithm"
2051 depends on MODULE_COMPRESS
2052 default MODULE_COMPRESS_GZIP
2053 help
2054 This determines which sort of compression will be used during
2055 'make modules_install'.
2056
2057 GZIP (default) and XZ are supported.
2058
2059config MODULE_COMPRESS_GZIP
2060 bool "GZIP"
2061
2062config MODULE_COMPRESS_XZ
2063 bool "XZ"
2064
2065endchoice
2066
Nicolas Pitredbacb0e2016-01-26 21:51:05 -05002067config TRIM_UNUSED_KSYMS
2068 bool "Trim unused exported kernel symbols"
2069 depends on MODULES && !UNUSED_SYMBOLS
2070 help
2071 The kernel and some modules make many symbols available for
2072 other modules to use via EXPORT_SYMBOL() and variants. Depending
2073 on the set of modules being selected in your kernel configuration,
2074 many of those exported symbols might never be used.
2075
2076 This option allows for unused exported symbols to be dropped from
2077 the build. In turn, this provides the compiler more opportunities
2078 (especially when using LTO) for optimizing the code and reducing
2079 binary size. This might have some security advantages as well.
2080
2081 If unsure say N.
2082
Robert P. J. Day0b0de142008-08-04 13:31:32 -04002083endif # MODULES
2084
Peter Zijlstra6c9692e2015-05-27 11:09:37 +09302085config MODULES_TREE_LOOKUP
2086 def_bool y
2087 depends on PERF_EVENTS || TRACING
2088
Rusty Russell98a79d62008-12-13 21:19:41 +10302089config INIT_ALL_POSSIBLE
2090 bool
2091 help
Rusty Russell5f054e32012-03-29 15:38:31 +10302092 Back when each arch used to define their own cpu_online_mask and
2093 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
Rusty Russell98a79d62008-12-13 21:19:41 +10302094 with all 1s, and others with all 0s. When they were centralised,
2095 it was better to provide this option than to break all the archs
Matt LaPlante692105b2009-01-26 11:12:25 +01002096 and have several arch maintainers pursuing me down dark alleys.
Rusty Russell98a79d62008-12-13 21:19:41 +10302097
Jens Axboe3a65dfe2005-11-04 08:43:35 +01002098source "block/Kconfig"
Avi Kivitye98c3202007-10-16 23:27:31 -07002099
2100config PREEMPT_NOTIFIERS
2101 bool
Paul E. McKenneye260be62008-01-25 21:08:24 +01002102
Steffen Klassert16295be2010-01-06 19:47:10 +11002103config PADATA
2104 depends on SMP
2105 bool
2106
Andi Kleen754b7b62012-10-04 17:11:27 -07002107# Can be selected by architectures with broken toolchains
2108# that get confused by correct const<->read_only section
2109# mappings
2110config BROKEN_RODATA
2111 bool
2112
David Howells4520c6a2012-09-21 23:31:13 +01002113config ASN1
2114 tristate
2115 help
2116 Build a simple ASN.1 grammar compiler that produces a bytecode output
2117 that can be interpreted by the ASN.1 stream decoder and used to
2118 inform it as to what tags are to be expected in a stream and what
2119 functions to call on what tags.
2120
Thomas Gleixner6beb0002009-11-09 15:21:34 +00002121source "kernel/Kconfig.locks"