]> git.proxmox.com Git - mirror_ubuntu-eoan-kernel.git/blob - kernel/power/Kconfig
docs: power: convert docs to ReST and rename to *.rst
[mirror_ubuntu-eoan-kernel.git] / kernel / power / Kconfig
1 config SUSPEND
2 bool "Suspend to RAM and standby"
3 depends on ARCH_SUSPEND_POSSIBLE
4 default y
5 ---help---
6 Allow the system to enter sleep states in which main memory is
7 powered and thus its contents are preserved, such as the
8 suspend-to-RAM state (e.g. the ACPI S3 state).
9
10 config SUSPEND_FREEZER
11 bool "Enable freezer for suspend to RAM/standby" \
12 if ARCH_WANTS_FREEZER_CONTROL || BROKEN
13 depends on SUSPEND
14 default y
15 help
16 This allows you to turn off the freezer for suspend. If this is
17 done, no tasks are frozen for suspend to RAM/standby.
18
19 Turning OFF this setting is NOT recommended! If in doubt, say Y.
20
21 config SUSPEND_SKIP_SYNC
22 bool "Skip kernel's sys_sync() on suspend to RAM/standby"
23 depends on SUSPEND
24 depends on EXPERT
25 help
26 Skip the kernel sys_sync() before freezing user processes.
27 Some systems prefer not to pay this cost on every invocation
28 of suspend, or they are content with invoking sync() from
29 user-space before invoking suspend. Say Y if that's your case.
30
31 config HIBERNATE_CALLBACKS
32 bool
33
34 config HIBERNATION
35 bool "Hibernation (aka 'suspend to disk')"
36 depends on SWAP && ARCH_HIBERNATION_POSSIBLE
37 select HIBERNATE_CALLBACKS
38 select LZO_COMPRESS
39 select LZO_DECOMPRESS
40 select CRC32
41 ---help---
42 Enable the suspend to disk (STD) functionality, which is usually
43 called "hibernation" in user interfaces. STD checkpoints the
44 system and powers it off; and restores that checkpoint on reboot.
45
46 You can suspend your machine with 'echo disk > /sys/power/state'
47 after placing resume=/dev/swappartition on the kernel command line
48 in your bootloader's configuration file.
49
50 Alternatively, you can use the additional userland tools available
51 from <http://suspend.sf.net>.
52
53 In principle it does not require ACPI or APM, although for example
54 ACPI will be used for the final steps when it is available. One
55 of the reasons to use software suspend is that the firmware hooks
56 for suspend states like suspend-to-RAM (STR) often don't work very
57 well with Linux.
58
59 It creates an image which is saved in your active swap. Upon the next
60 boot, pass the 'resume=/dev/swappartition' argument to the kernel to
61 have it detect the saved image, restore memory state from it, and
62 continue to run as before. If you do not want the previous state to
63 be reloaded, then use the 'noresume' kernel command line argument.
64 Note, however, that fsck will be run on your filesystems and you will
65 need to run mkswap against the swap partition used for the suspend.
66
67 It also works with swap files to a limited extent (for details see
68 <file:Documentation/power/swsusp-and-swap-files.rst>).
69
70 Right now you may boot without resuming and resume later but in the
71 meantime you cannot use the swap partition(s)/file(s) involved in
72 suspending. Also in this case you must not use the filesystems
73 that were mounted before the suspend. In particular, you MUST NOT
74 MOUNT any journaled filesystems mounted before the suspend or they
75 will get corrupted in a nasty way.
76
77 For more information take a look at <file:Documentation/power/swsusp.rst>.
78
79 config ARCH_SAVE_PAGE_KEYS
80 bool
81
82 config PM_STD_PARTITION
83 string "Default resume partition"
84 depends on HIBERNATION
85 default ""
86 ---help---
87 The default resume partition is the partition that the suspend-
88 to-disk implementation will look for a suspended disk image.
89
90 The partition specified here will be different for almost every user.
91 It should be a valid swap partition (at least for now) that is turned
92 on before suspending.
93
94 The partition specified can be overridden by specifying:
95
96 resume=/dev/<other device>
97
98 which will set the resume partition to the device specified.
99
100 Note there is currently not a way to specify which device to save the
101 suspended image to. It will simply pick the first available swap
102 device.
103
104 config PM_SLEEP
105 def_bool y
106 depends on SUSPEND || HIBERNATE_CALLBACKS
107 select PM
108 select SRCU
109
110 config PM_SLEEP_SMP
111 def_bool y
112 depends on SMP
113 depends on ARCH_SUSPEND_POSSIBLE || ARCH_HIBERNATION_POSSIBLE
114 depends on PM_SLEEP
115 select HOTPLUG_CPU
116
117 config PM_SLEEP_SMP_NONZERO_CPU
118 def_bool y
119 depends on PM_SLEEP_SMP
120 depends on ARCH_SUSPEND_NONZERO_CPU
121 ---help---
122 If an arch can suspend (for suspend, hibernate, kexec, etc) on a
123 non-zero numbered CPU, it may define ARCH_SUSPEND_NONZERO_CPU. This
124 will allow nohz_full mask to include CPU0.
125
126 config PM_AUTOSLEEP
127 bool "Opportunistic sleep"
128 depends on PM_SLEEP
129 default n
130 ---help---
131 Allow the kernel to trigger a system transition into a global sleep
132 state automatically whenever there are no active wakeup sources.
133
134 config PM_WAKELOCKS
135 bool "User space wakeup sources interface"
136 depends on PM_SLEEP
137 default n
138 ---help---
139 Allow user space to create, activate and deactivate wakeup source
140 objects with the help of a sysfs-based interface.
141
142 config PM_WAKELOCKS_LIMIT
143 int "Maximum number of user space wakeup sources (0 = no limit)"
144 range 0 100000
145 default 100
146 depends on PM_WAKELOCKS
147
148 config PM_WAKELOCKS_GC
149 bool "Garbage collector for user space wakeup sources"
150 depends on PM_WAKELOCKS
151 default y
152
153 config PM
154 bool "Device power management core functionality"
155 ---help---
156 Enable functionality allowing I/O devices to be put into energy-saving
157 (low power) states, for example after a specified period of inactivity
158 (autosuspended), and woken up in response to a hardware-generated
159 wake-up event or a driver's request.
160
161 Hardware support is generally required for this functionality to work
162 and the bus type drivers of the buses the devices are on are
163 responsible for the actual handling of device suspend requests and
164 wake-up events.
165
166 config PM_DEBUG
167 bool "Power Management Debug Support"
168 depends on PM
169 ---help---
170 This option enables various debugging support in the Power Management
171 code. This is helpful when debugging and reporting PM bugs, like
172 suspend support.
173
174 config PM_ADVANCED_DEBUG
175 bool "Extra PM attributes in sysfs for low-level debugging/testing"
176 depends on PM_DEBUG
177 ---help---
178 Add extra sysfs attributes allowing one to access some Power Management
179 fields of device objects from user space. If you are not a kernel
180 developer interested in debugging/testing Power Management, say "no".
181
182 config PM_TEST_SUSPEND
183 bool "Test suspend/resume and wakealarm during bootup"
184 depends on SUSPEND && PM_DEBUG && RTC_CLASS=y
185 ---help---
186 This option will let you suspend your machine during bootup, and
187 make it wake up a few seconds later using an RTC wakeup alarm.
188 Enable this with a kernel parameter like "test_suspend=mem".
189
190 You probably want to have your system's RTC driver statically
191 linked, ensuring that it's available when this test runs.
192
193 config PM_SLEEP_DEBUG
194 def_bool y
195 depends on PM_DEBUG && PM_SLEEP
196
197 config DPM_WATCHDOG
198 bool "Device suspend/resume watchdog"
199 depends on PM_DEBUG && PSTORE && EXPERT
200 ---help---
201 Sets up a watchdog timer to capture drivers that are
202 locked up attempting to suspend/resume a device.
203 A detected lockup causes system panic with message
204 captured in pstore device for inspection in subsequent
205 boot session.
206
207 config DPM_WATCHDOG_TIMEOUT
208 int "Watchdog timeout in seconds"
209 range 1 120
210 default 120
211 depends on DPM_WATCHDOG
212
213 config PM_TRACE
214 bool
215 help
216 This enables code to save the last PM event point across
217 reboot. The architecture needs to support this, x86 for
218 example does by saving things in the RTC, see below.
219
220 The architecture specific code must provide the extern
221 functions from <linux/resume-trace.h> as well as the
222 <asm/resume-trace.h> header with a TRACE_RESUME() macro.
223
224 The way the information is presented is architecture-
225 dependent, x86 will print the information during a
226 late_initcall.
227
228 config PM_TRACE_RTC
229 bool "Suspend/resume event tracing"
230 depends on PM_SLEEP_DEBUG
231 depends on X86
232 select PM_TRACE
233 ---help---
234 This enables some cheesy code to save the last PM event point in the
235 RTC across reboots, so that you can debug a machine that just hangs
236 during suspend (or more commonly, during resume).
237
238 To use this debugging feature you should attempt to suspend the
239 machine, reboot it and then run
240
241 dmesg -s 1000000 | grep 'hash matches'
242
243 CAUTION: this option will cause your machine's real-time clock to be
244 set to an invalid time after a resume.
245
246 config APM_EMULATION
247 tristate "Advanced Power Management Emulation"
248 depends on SYS_SUPPORTS_APM_EMULATION
249 help
250 APM is a BIOS specification for saving power using several different
251 techniques. This is mostly useful for battery powered laptops with
252 APM compliant BIOSes. If you say Y here, the system time will be
253 reset after a RESUME operation, the /proc/apm device will provide
254 battery status information, and user-space programs will receive
255 notification of APM "events" (e.g. battery status change).
256
257 In order to use APM, you will need supporting software. For location
258 and more information, read <file:Documentation/power/apm-acpi.rst>
259 and the Battery Powered Linux mini-HOWTO, available from
260 <http://www.tldp.org/docs.html#howto>.
261
262 This driver does not spin down disk drives (see the hdparm(8)
263 manpage ("man 8 hdparm") for that), and it doesn't turn off
264 VESA-compliant "green" monitors.
265
266 Generally, if you don't have a battery in your machine, there isn't
267 much point in using this driver and you should say N. If you get
268 random kernel OOPSes or reboots that don't seem to be related to
269 anything, try disabling/enabling this option (or disabling/enabling
270 APM in your BIOS).
271
272 config PM_CLK
273 def_bool y
274 depends on PM && HAVE_CLK
275
276 config PM_GENERIC_DOMAINS
277 bool
278 depends on PM
279
280 config WQ_POWER_EFFICIENT_DEFAULT
281 bool "Enable workqueue power-efficient mode by default"
282 depends on PM
283 default n
284 help
285 Per-cpu workqueues are generally preferred because they show
286 better performance thanks to cache locality; unfortunately,
287 per-cpu workqueues tend to be more power hungry than unbound
288 workqueues.
289
290 Enabling workqueue.power_efficient kernel parameter makes the
291 per-cpu workqueues which were observed to contribute
292 significantly to power consumption unbound, leading to measurably
293 lower power usage at the cost of small performance overhead.
294
295 This config option determines whether workqueue.power_efficient
296 is enabled by default.
297
298 If in doubt, say N.
299
300 config PM_GENERIC_DOMAINS_SLEEP
301 def_bool y
302 depends on PM_SLEEP && PM_GENERIC_DOMAINS
303
304 config PM_GENERIC_DOMAINS_OF
305 def_bool y
306 depends on PM_GENERIC_DOMAINS && OF
307
308 config CPU_PM
309 bool
310
311 config ENERGY_MODEL
312 bool "Energy Model for CPUs"
313 depends on SMP
314 depends on CPU_FREQ
315 default n
316 help
317 Several subsystems (thermal and/or the task scheduler for example)
318 can leverage information about the energy consumed by CPUs to make
319 smarter decisions. This config option enables the framework from
320 which subsystems can access the energy models.
321
322 The exact usage of the energy model is subsystem-dependent.
323
324 If in doubt, say N.