3 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
5 The /sys/power directory will contain files that will
6 provide a unified interface to the power management
11 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
13 The /sys/power/state file controls system sleep states.
14 Reading from this file returns the available sleep state
15 labels, which may be "mem" (suspend), "standby" (power-on
16 suspend), "freeze" (suspend-to-idle) and "disk" (hibernation).
18 Writing one of the above strings to this file causes the system
19 to transition into the corresponding state, if available.
21 See Documentation/admin-guide/pm/sleep-states.rst for more
24 What: /sys/power/mem_sleep
26 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
28 The /sys/power/mem_sleep file controls the operating mode of
29 system suspend. Reading from it returns the available modes
30 as "s2idle" (always present), "shallow" and "deep" (present if
31 supported). The mode that will be used on subsequent attempts
32 to suspend the system (by writing "mem" to the /sys/power/state
33 file described above) is enclosed in square brackets.
35 Writing one of the above strings to this file causes the mode
36 represented by it to be used on subsequent attempts to suspend
39 See Documentation/admin-guide/pm/sleep-states.rst for more
44 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
46 The /sys/power/disk file controls the operating mode of the
47 suspend-to-disk mechanism. Reading from this file returns
48 the name of the method by which the system will be put to
49 sleep on the next suspend. There are four methods supported:
51 'firmware' - means that the memory image will be saved to disk
52 by some firmware, in which case we also assume that the
53 firmware will handle the system suspend.
55 'platform' - the memory image will be saved by the kernel and
56 the system will be put to sleep by the platform driver (e.g.
57 ACPI or other PM registers).
59 'shutdown' - the memory image will be saved by the kernel and
60 the system will be powered off.
62 'reboot' - the memory image will be saved by the kernel and
63 the system will be rebooted.
65 Additionally, /sys/power/disk can be used to turn on one of the
66 two testing modes of the suspend-to-disk mechanism: 'testproc'
67 or 'test'. If the suspend-to-disk mechanism is in the
68 'testproc' mode, writing 'disk' to /sys/power/state will cause
69 the kernel to disable nonboot CPUs and freeze tasks, wait for 5
70 seconds, unfreeze tasks and enable nonboot CPUs. If it is in
71 the 'test' mode, writing 'disk' to /sys/power/state will cause
72 the kernel to disable nonboot CPUs and freeze tasks, shrink
73 memory, suspend devices, wait for 5 seconds, resume devices,
74 unfreeze tasks and enable nonboot CPUs. Then, we are able to
75 look in the log messages and work out, for example, which code
76 is being slow and which device drivers are misbehaving.
78 The suspend-to-disk method may be chosen by writing to this
79 file one of the accepted strings:
88 It will only change to 'firmware' or 'platform' if the system
91 What: /sys/power/image_size
93 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
95 The /sys/power/image_size file controls the size of the image
96 created by the suspend-to-disk mechanism. It can be written a
97 string representing a non-negative integer that will be used
98 as an upper limit of the image size, in bytes. The kernel's
99 suspend-to-disk code will do its best to ensure the image size
100 will not exceed this number. However, if it turns out to be
101 impossible, the kernel will try to suspend anyway using the
102 smallest image possible. In particular, if "0" is written to
103 this file, the suspend image will be as small as possible.
105 Reading from this file will display the current image size
106 limit, which is set to around 2/5 of available RAM by default.
108 What: /sys/power/pm_trace
110 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
112 The /sys/power/pm_trace file controls the code which saves the
113 last PM event point in the RTC across reboots, so that you can
114 debug a machine that just hangs during suspend (or more
115 commonly, during resume). Namely, the RTC is only used to save
116 the last PM event point if this file contains '1'. Initially
117 it contains '0' which may be changed to '1' by writing a
118 string representing a nonzero integer into it.
120 To use this debugging feature you should attempt to suspend
121 the machine, then reboot it and run::
123 dmesg -s 1000000 | grep 'hash matches'
125 If you do not get any matches (or they appear to be false
126 positives), it is possible that the last PM event point
127 referred to a device created by a loadable kernel module. In
128 this case cat /sys/power/pm_trace_dev_match (see below) after
129 your system is started up and the kernel modules are loaded.
131 CAUTION: Using it will cause your machine's real-time (CMOS)
132 clock to be set to a random invalid time after a resume.
134 What; /sys/power/pm_trace_dev_match
136 Contact: James Hogan <jhogan@kernel.org>
138 The /sys/power/pm_trace_dev_match file contains the name of the
139 device associated with the last PM event point saved in the RTC
140 across reboots when pm_trace has been used. More precisely it
141 contains the list of current devices (including those
142 registered by loadable kernel modules since boot) which match
143 the device hash in the RTC at boot, with a newline after each
146 The advantage of this file over the hash matches printed to the
147 kernel log (see /sys/power/pm_trace), is that it includes
148 devices created after boot by loadable kernel modules.
150 Due to the small hash size necessary to fit in the RTC, it is
151 possible that more than one device matches the hash, in which
152 case further investigation is required to determine which
153 device is causing the problem. Note that genuine RTC clock
154 values (such as when pm_trace has not been used), can still
155 match a device and output it's name here.
157 What: /sys/power/pm_async
159 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
161 The /sys/power/pm_async file controls the switch allowing the
162 user space to enable or disable asynchronous suspend and resume
163 of devices. If enabled, this feature will cause some device
164 drivers' suspend and resume callbacks to be executed in parallel
165 with each other and with the main suspend thread. It is enabled
166 if this file contains "1", which is the default. It may be
167 disabled by writing "0" to this file, in which case all devices
168 will be suspended and resumed synchronously.
170 What: /sys/power/wakeup_count
172 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
174 The /sys/power/wakeup_count file allows user space to put the
175 system into a sleep state while taking into account the
176 concurrent arrival of wakeup events. Reading from it returns
177 the current number of registered wakeup events and it blocks if
178 some wakeup events are being processed at the time the file is
179 read from. Writing to it will only succeed if the current
180 number of wakeup events is equal to the written value and, if
181 successful, will make the kernel abort a subsequent transition
182 to a sleep state if any wakeup events are reported after the
185 What: /sys/power/reserved_size
187 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
189 The /sys/power/reserved_size file allows user space to control
190 the amount of memory reserved for allocations made by device
191 drivers during the "device freeze" stage of hibernation. It can
192 be written a string representing a non-negative integer that
193 will be used as the amount of memory to reserve for allocations
194 made by device drivers' "freeze" callbacks, in bytes.
196 Reading from this file will display the current value, which is
197 set to 1 MB by default.
199 What: /sys/power/autosleep
201 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
203 The /sys/power/autosleep file can be written one of the strings
204 returned by reads from /sys/power/state. If that happens, a
205 work item attempting to trigger a transition of the system to
206 the sleep state represented by that string is queued up. This
207 attempt will only succeed if there are no active wakeup sources
208 in the system at that time. After every execution, regardless
209 of whether or not the attempt to put the system to sleep has
210 succeeded, the work item requeues itself until user space
211 writes "off" to /sys/power/autosleep.
213 Reading from this file causes the last string successfully
214 written to it to be returned.
216 What: /sys/power/wake_lock
218 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
220 The /sys/power/wake_lock file allows user space to create
221 wakeup source objects and activate them on demand (if one of
222 those wakeup sources is active, reads from the
223 /sys/power/wakeup_count file block or return false). When a
224 string without white space is written to /sys/power/wake_lock,
225 it will be assumed to represent a wakeup source name. If there
226 is a wakeup source object with that name, it will be activated
227 (unless active already). Otherwise, a new wakeup source object
228 will be registered, assigned the given name and activated.
229 If a string written to /sys/power/wake_lock contains white
230 space, the part of the string preceding the white space will be
231 regarded as a wakeup source name and handled as descrived above.
232 The other part of the string will be regarded as a timeout (in
233 nanoseconds) such that the wakeup source will be automatically
234 deactivated after it has expired. The timeout, if present, is
235 set regardless of the current state of the wakeup source object
238 Reads from this file return a string consisting of the names of
239 wakeup sources created with the help of it that are active at
240 the moment, separated with spaces.
243 What: /sys/power/wake_unlock
245 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
247 The /sys/power/wake_unlock file allows user space to deactivate
248 wakeup sources created with the help of /sys/power/wake_lock.
249 When a string is written to /sys/power/wake_unlock, it will be
250 assumed to represent the name of a wakeup source to deactivate.
252 If a wakeup source object of that name exists and is active at
253 the moment, it will be deactivated.
255 Reads from this file return a string consisting of the names of
256 wakeup sources created with the help of /sys/power/wake_lock
257 that are inactive at the moment, separated with spaces.
259 What: /sys/power/pm_print_times
261 Contact: Sameer Nanda <snanda@chromium.org>
263 The /sys/power/pm_print_times file allows user space to
264 control whether the time taken by devices to suspend and
265 resume is printed. These prints are useful for hunting down
266 devices that take too long to suspend or resume.
268 Writing a "1" enables this printing while writing a "0"
269 disables it. The default value is "0". Reading from this file
270 will display the current value.
272 What: /sys/power/pm_wakeup_irq
274 Contact: Alexandra Yates <alexandra.yates@linux.intel.org>
276 The /sys/power/pm_wakeup_irq file reports to user space the IRQ
277 number of the first wakeup interrupt (that is, the first
278 interrupt from an IRQ line armed for system wakeup) seen by the
279 kernel during the most recent system suspend/resume cycle.
281 This output is useful for system wakeup diagnostics of spurious
284 What: /sys/power/pm_debug_messages
286 Contact: Rafael J. Wysocki <rjw@rjwysocki.net>
288 The /sys/power/pm_debug_messages file controls the printing
289 of debug messages from the system suspend/hiberbation
290 infrastructure to the kernel log.
292 Writing a "1" to this file enables the debug messages and
293 writing a "0" (default) to it disables them. Reads from
294 this file return the current value.
296 What: /sys/power/resume_offset
298 Contact: Mario Limonciello <mario.limonciello@dell.com>
300 This file is used for telling the kernel an offset into a disk
301 to use when hibernating the system such as with a swap file.
303 Reads from this file will display the current offset
304 the kernel will be using on the next hibernation
307 Using this sysfs file will override any values that were
308 set using the kernel command line for disk offset.
310 What: /sys/power/suspend_stats
312 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
314 The /sys/power/suspend_stats directory contains suspend related
317 What: /sys/power/suspend_stats/success
319 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
321 The /sys/power/suspend_stats/success file contains the number
322 of times entering system sleep state succeeded.
324 What: /sys/power/suspend_stats/fail
326 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
328 The /sys/power/suspend_stats/fail file contains the number
329 of times entering system sleep state failed.
331 What: /sys/power/suspend_stats/failed_freeze
333 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
335 The /sys/power/suspend_stats/failed_freeze file contains the
336 number of times freezing processes failed.
338 What: /sys/power/suspend_stats/failed_prepare
340 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
342 The /sys/power/suspend_stats/failed_prepare file contains the
343 number of times preparing all non-sysdev devices for
344 a system PM transition failed.
346 What: /sys/power/suspend_stats/failed_resume
348 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
350 The /sys/power/suspend_stats/failed_resume file contains the
351 number of times executing "resume" callbacks of
352 non-sysdev devices failed.
354 What: /sys/power/suspend_stats/failed_resume_early
356 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
358 The /sys/power/suspend_stats/failed_resume_early file contains
359 the number of times executing "early resume" callbacks
362 What: /sys/power/suspend_stats/failed_resume_noirq
364 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
366 The /sys/power/suspend_stats/failed_resume_noirq file contains
367 the number of times executing "noirq resume" callbacks
370 What: /sys/power/suspend_stats/failed_suspend
372 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
374 The /sys/power/suspend_stats/failed_suspend file contains
375 the number of times executing "suspend" callbacks
376 of all non-sysdev devices failed.
378 What: /sys/power/suspend_stats/failed_suspend_late
380 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
382 The /sys/power/suspend_stats/failed_suspend_late file contains
383 the number of times executing "late suspend" callbacks
384 of all devices failed.
386 What: /sys/power/suspend_stats/failed_suspend_noirq
388 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
390 The /sys/power/suspend_stats/failed_suspend_noirq file contains
391 the number of times executing "noirq suspend" callbacks
392 of all devices failed.
394 What: /sys/power/suspend_stats/last_failed_dev
396 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
398 The /sys/power/suspend_stats/last_failed_dev file contains
399 the last device for which a suspend/resume callback failed.
401 What: /sys/power/suspend_stats/last_failed_errno
403 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
405 The /sys/power/suspend_stats/last_failed_errno file contains
406 the errno of the last failed attempt at entering
409 What: /sys/power/suspend_stats/last_failed_step
411 Contact: Kalesh Singh <kaleshsingh96@gmail.com>
413 The /sys/power/suspend_stats/last_failed_step file contains
414 the last failed step in the suspend/resume path.
416 What: /sys/power/sync_on_suspend
418 Contact: Jonas Meurer <jonas@freesources.org>
420 This file controls whether or not the kernel will sync()
421 filesystems during system suspend (after freezing user space
422 and before suspending devices).
424 Writing a "1" to this file enables the sync() and writing a "0"
425 disables it. Reads from the file return the current value.
426 The default is "1" if the build-time "SUSPEND_SKIP_SYNC" config
427 flag is unset, or "0" otherwise.