1What:		/sys/power/
2Date:		August 2006
3Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
4Description:
5		The /sys/power directory will contain files that will
6		provide a unified interface to the power management
7		subsystem.
8
9What:		/sys/power/state
10Date:		November 2016
11Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
12Description:
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).
17
18		Writing one of the above strings to this file causes the system
19		to transition into the corresponding state, if available.
20
21		See Documentation/power/states.txt for more information.
22
23What:		/sys/power/mem_sleep
24Date:		November 2016
25Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
26Description:
27		The /sys/power/mem_sleep file controls the operating mode of
28		system suspend.  Reading from it returns the available modes
29		as "s2idle" (always present), "shallow" and "deep" (present if
30		supported).  The mode that will be used on subsequent attempts
31		to suspend the system (by writing "mem" to the /sys/power/state
32		file described above) is enclosed in square brackets.
33
34		Writing one of the above strings to this file causes the mode
35		represented by it to be used on subsequent attempts to suspend
36		the system.
37
38		See Documentation/power/states.txt for more information.
39
40What:		/sys/power/disk
41Date:		September 2006
42Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
43Description:
44		The /sys/power/disk file controls the operating mode of the
45		suspend-to-disk mechanism.  Reading from this file returns
46		the name of the method by which the system will be put to
47		sleep on the next suspend.  There are four methods supported:
48		'firmware' - means that the memory image will be saved to disk
49		by some firmware, in which case we also assume that the
50		firmware will handle the system suspend.
51		'platform' - the memory image will be saved by the kernel and
52		the system will be put to sleep by the platform driver (e.g.
53		ACPI or other PM registers).
54		'shutdown' - the memory image will be saved by the kernel and
55		the system will be powered off.
56		'reboot' - the memory image will be saved by the kernel and
57		the system will be rebooted.
58
59		Additionally, /sys/power/disk can be used to turn on one of the
60		two testing modes of the suspend-to-disk mechanism: 'testproc'
61		or 'test'.  If the suspend-to-disk mechanism is in the
62		'testproc' mode, writing 'disk' to /sys/power/state will cause
63		the kernel to disable nonboot CPUs and freeze tasks, wait for 5
64		seconds, unfreeze tasks and enable nonboot CPUs.  If it is in
65		the 'test' mode, writing 'disk' to /sys/power/state will cause
66		the kernel to disable nonboot CPUs and freeze tasks, shrink
67		memory, suspend devices, wait for 5 seconds, resume devices,
68		unfreeze tasks and enable nonboot CPUs.  Then, we are able to
69		look in the log messages and work out, for example, which code
70		is being slow and which device drivers are misbehaving.
71
72		The suspend-to-disk method may be chosen by writing to this
73		file one of the accepted strings:
74
75		'firmware'
76		'platform'
77		'shutdown'
78		'reboot'
79		'testproc'
80		'test'
81
82		It will only change to 'firmware' or 'platform' if the system
83		supports that.
84
85What:		/sys/power/image_size
86Date:		August 2006
87Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
88Description:
89		The /sys/power/image_size file controls the size of the image
90		created by the suspend-to-disk mechanism.  It can be written a
91		string representing a non-negative integer that will be used
92		as an upper limit of the image size, in bytes.  The kernel's
93		suspend-to-disk code will do its best to ensure the image size
94		will not exceed this number.  However, if it turns out to be
95		impossible, the kernel will try to suspend anyway using the
96		smallest image possible.  In particular, if "0" is written to
97		this file, the suspend image will be as small as possible.
98
99		Reading from this file will display the current image size
100		limit, which is set to 500 MB by default.
101
102What:		/sys/power/pm_trace
103Date:		August 2006
104Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
105Description:
106		The /sys/power/pm_trace file controls the code which saves the
107		last PM event point in the RTC across reboots, so that you can
108		debug a machine that just hangs during suspend (or more
109		commonly, during resume).  Namely, the RTC is only used to save
110		the last PM event point if this file contains '1'.  Initially
111		it contains '0' which may be changed to '1' by writing a
112		string representing a nonzero integer into it.
113
114		To use this debugging feature you should attempt to suspend
115		the machine, then reboot it and run
116
117		dmesg -s 1000000 | grep 'hash matches'
118
119		If you do not get any matches (or they appear to be false
120		positives), it is possible that the last PM event point
121		referred to a device created by a loadable kernel module.  In
122		this case cat /sys/power/pm_trace_dev_match (see below) after
123		your system is started up and the kernel modules are loaded.
124
125		CAUTION: Using it will cause your machine's real-time (CMOS)
126		clock to be set to a random invalid time after a resume.
127
128What;		/sys/power/pm_trace_dev_match
129Date:		October 2010
130Contact:	James Hogan <james@albanarts.com>
131Description:
132		The /sys/power/pm_trace_dev_match file contains the name of the
133		device associated with the last PM event point saved in the RTC
134		across reboots when pm_trace has been used.  More precisely it
135		contains the list of current devices (including those
136		registered by loadable kernel modules since boot) which match
137		the device hash in the RTC at boot, with a newline after each
138		one.
139
140		The advantage of this file over the hash matches printed to the
141		kernel log (see /sys/power/pm_trace), is that it includes
142		devices created after boot by loadable kernel modules.
143
144		Due to the small hash size necessary to fit in the RTC, it is
145		possible that more than one device matches the hash, in which
146		case further investigation is required to determine which
147		device is causing the problem.  Note that genuine RTC clock
148		values (such as when pm_trace has not been used), can still
149		match a device and output it's name here.
150
151What:		/sys/power/pm_async
152Date:		January 2009
153Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
154Description:
155		The /sys/power/pm_async file controls the switch allowing the
156		user space to enable or disable asynchronous suspend and resume
157		of devices.  If enabled, this feature will cause some device
158		drivers' suspend and resume callbacks to be executed in parallel
159		with each other and with the main suspend thread.  It is enabled
160		if this file contains "1", which is the default.  It may be
161		disabled by writing "0" to this file, in which case all devices
162		will be suspended and resumed synchronously.
163
164What:		/sys/power/wakeup_count
165Date:		July 2010
166Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
167Description:
168		The /sys/power/wakeup_count file allows user space to put the
169		system into a sleep state while taking into account the
170		concurrent arrival of wakeup events.  Reading from it returns
171		the current number of registered wakeup events and it blocks if
172		some wakeup events are being processed at the time the file is
173		read from.  Writing to it will only succeed if the current
174		number of wakeup events is equal to the written value and, if
175		successful, will make the kernel abort a subsequent transition
176		to a sleep state if any wakeup events are reported after the
177		write has returned.
178
179What:		/sys/power/reserved_size
180Date:		May 2011
181Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
182Description:
183		The /sys/power/reserved_size file allows user space to control
184		the amount of memory reserved for allocations made by device
185		drivers during the "device freeze" stage of hibernation.  It can
186		be written a string representing a non-negative integer that
187		will be used as the amount of memory to reserve for allocations
188		made by device drivers' "freeze" callbacks, in bytes.
189
190		Reading from this file will display the current value, which is
191		set to 1 MB by default.
192
193What:		/sys/power/autosleep
194Date:		April 2012
195Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
196Description:
197		The /sys/power/autosleep file can be written one of the strings
198		returned by reads from /sys/power/state.  If that happens, a
199		work item attempting to trigger a transition of the system to
200		the sleep state represented by that string is queued up.  This
201		attempt will only succeed if there are no active wakeup sources
202		in the system at that time.  After every execution, regardless
203		of whether or not the attempt to put the system to sleep has
204		succeeded, the work item requeues itself until user space
205		writes "off" to /sys/power/autosleep.
206
207		Reading from this file causes the last string successfully
208		written to it to be returned.
209
210What:		/sys/power/wake_lock
211Date:		February 2012
212Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
213Description:
214		The /sys/power/wake_lock file allows user space to create
215		wakeup source objects and activate them on demand (if one of
216		those wakeup sources is active, reads from the
217		/sys/power/wakeup_count file block or return false).  When a
218		string without white space is written to /sys/power/wake_lock,
219		it will be assumed to represent a wakeup source name.  If there
220		is a wakeup source object with that name, it will be activated
221		(unless active already).  Otherwise, a new wakeup source object
222		will be registered, assigned the given name and activated.
223		If a string written to /sys/power/wake_lock contains white
224		space, the part of the string preceding the white space will be
225		regarded as a wakeup source name and handled as descrived above.
226		The other part of the string will be regarded as a timeout (in
227		nanoseconds) such that the wakeup source will be automatically
228		deactivated after it has expired.  The timeout, if present, is
229		set regardless of the current state of the wakeup source object
230		in question.
231
232		Reads from this file return a string consisting of the names of
233		wakeup sources created with the help of it that are active at
234		the moment, separated with spaces.
235
236
237What:		/sys/power/wake_unlock
238Date:		February 2012
239Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
240Description:
241		The /sys/power/wake_unlock file allows user space to deactivate
242		wakeup sources created with the help of /sys/power/wake_lock.
243		When a string is written to /sys/power/wake_unlock, it will be
244		assumed to represent the name of a wakeup source to deactivate.
245		If a wakeup source object of that name exists and is active at
246		the moment, it will be deactivated.
247
248		Reads from this file return a string consisting of the names of
249		wakeup sources created with the help of /sys/power/wake_lock
250		that are inactive at the moment, separated with spaces.
251
252What:		/sys/power/pm_print_times
253Date:		May 2012
254Contact:	Sameer Nanda <snanda@chromium.org>
255Description:
256		The /sys/power/pm_print_times file allows user space to
257		control whether the time taken by devices to suspend and
258		resume is printed.  These prints are useful for hunting down
259		devices that take too long to suspend or resume.
260
261		Writing a "1" enables this printing while writing a "0"
262		disables it.  The default value is "0".  Reading from this file
263		will display the current value.
264
265What:		/sys/power/pm_wakeup_irq
266Date:		April 2015
267Contact:	Alexandra Yates <alexandra.yates@linux.intel.org>
268Description:
269		The /sys/power/pm_wakeup_irq file reports to user space the IRQ
270		number of the first wakeup interrupt (that is, the first
271		interrupt from an IRQ line armed for system wakeup) seen by the
272		kernel during the most recent system suspend/resume cycle.
273
274		This output is useful for system wakeup diagnostics of spurious
275		wakeup interrupts.
276