Documentation/power/runtime_pm.txt

   1 Run-time Power Management Framework for I/O Devices
   2
   3 (C) 2009 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc.
   4
   5 1. Introduction
   6
   7 Support for run-time power management (run-time PM) of I/O devices is provided
   8 at the power management core (PM core) level by means of:
   9
  10 * The power management workqueue pm_wq in which bus types and device drivers can
  11   put their PM-related work items.  It is strongly recommended that pm_wq be
  12   used for queuing all work items related to run-time PM, because this allows
  13   them to be synchronized with system-wide power transitions (suspend to RAM,
  14   hibernation and resume from system sleep states).  pm_wq is declared in
  15   include/linux/pm_runtime.h and defined in kernel/power/main.c.
  16
  17 * A number of run-time PM fields in the 'power' member of 'struct device' (which
  18   is of the type 'struct dev_pm_info', defined in include/linux/pm.h) that can
  19   be used for synchronizing run-time PM operations with one another.
  20
  21 * Three device run-time PM callbacks in 'struct dev_pm_ops' (defined in
  22   include/linux/pm.h).
  23
  24 * A set of helper functions defined in drivers/base/power/runtime.c that can be
  25   used for carrying out run-time PM operations in such a way that the
  26   synchronization between them is taken care of by the PM core.  Bus types and
  27   device drivers are encouraged to use these functions.
  28
  29 The run-time PM callbacks present in 'struct dev_pm_ops', the device run-time PM
  30 fields of 'struct dev_pm_info' and the core helper functions provided for
  31 run-time PM are described below.
  32
  33 2. Device Run-time PM Callbacks
  34
  35 There are three device run-time PM callbacks defined in 'struct dev_pm_ops':
  36
  37 struct dev_pm_ops {
  38         ...
  39         int (*runtime_suspend)(struct device *dev);
  40         int (*runtime_resume)(struct device *dev);
  41         void (*runtime_idle)(struct device *dev);
  42         ...
  43 };
  44
  45 The ->runtime_suspend() callback is executed by the PM core for the bus type of
  46 the device being suspended.  The bus type's callback is then _entirely_
  47 _responsible_ for handling the device as appropriate, which may, but need not
  48 include executing the device driver's own ->runtime_suspend() callback (from the
  49 PM core's point of view it is not necessary to implement a ->runtime_suspend()
  50 callback in a device driver as long as the bus type's ->runtime_suspend() knows
  51 what to do to handle the device).
  52
  53   * Once the bus type's ->runtime_suspend() callback has completed successfully
  54     for given device, the PM core regards the device as suspended, which need
  55     not mean that the device has been put into a low power state.  It is
  56     supposed to mean, however, that the device will not process data and will
  57     not communicate with the CPU(s) and RAM until its bus type's
  58     ->runtime_resume() callback is executed for it.  The run-time PM status of
  59     a device after successful execution of its bus type's ->runtime_suspend()
  60     callback is 'suspended'.
  61
  62   * If the bus type's ->runtime_suspend() callback returns -EBUSY or -EAGAIN,
  63     the device's run-time PM status is supposed to be 'active', which means that
  64     the device _must_ be fully operational afterwards.
  65
  66   * If the bus type's ->runtime_suspend() callback returns an error code
  67     different from -EBUSY or -EAGAIN, the PM core regards this as a fatal
  68     error and will refuse to run the helper functions described in Section 4
  69     for the device, until the status of it is directly set either to 'active'
  70     or to 'suspended' (the PM core provides special helper functions for this
  71     purpose).
  72
  73 In particular, if the driver requires remote wakeup capability for proper
  74 functioning and device_may_wakeup() returns 'false' for the device, then
  75 ->runtime_suspend() should return -EBUSY.  On the other hand, if
  76 device_may_wakeup() returns 'true' for the device and the device is put
  77 into a low power state during the execution of its bus type's
  78 ->runtime_suspend(), it is expected that remote wake-up (i.e. hardware mechanism
  79 allowing the device to request a change of its power state, such as PCI PME)
  80 will be enabled for the device.  Generally, remote wake-up should be enabled
  81 for all input devices put into a low power state at run time.
  82
  83 The ->runtime_resume() callback is executed by the PM core for the bus type of
  84 the device being woken up.  The bus type's callback is then _entirely_
  85 _responsible_ for handling the device as appropriate, which may, but need not
  86 include executing the device driver's own ->runtime_resume() callback (from the
  87 PM core's point of view it is not necessary to implement a ->runtime_resume()
  88 callback in a device driver as long as the bus type's ->runtime_resume() knows
  89 what to do to handle the device).
  90
  91   * Once the bus type's ->runtime_resume() callback has completed successfully,
  92     the PM core regards the device as fully operational, which means that the
  93     device _must_ be able to complete I/O operations as needed.  The run-time
  94     PM status of the device is then 'active'.
  95
  96   * If the bus type's ->runtime_resume() callback returns an error code, the PM
  97     core regards this as a fatal error and will refuse to run the helper
  98     functions described in Section 4 for the device, until its status is
  99     directly set either to 'active' or to 'suspended' (the PM core provides
 100     special helper functions for this purpose).
 101
 102 The ->runtime_idle() callback is executed by the PM core for the bus type of
 103 given device whenever the device appears to be idle, which is indicated to the
 104 PM core by two counters, the device's usage counter and the counter of 'active'
 105 children of the device.
 106
 107   * If any of these counters is decreased using a helper function provided by
 108     the PM core and it turns out to be equal to zero, the other counter is
 109     checked.  If that counter also is equal to zero, the PM core executes the
 110     device bus type's ->runtime_idle() callback (with the device as an
 111     argument).
 112
 113 The action performed by a bus type's ->runtime_idle() callback is totally
 114 dependent on the bus type in question, but the expected and recommended action
 115 is to check if the device can be suspended (i.e. if all of the conditions
 116 necessary for suspending the device are satisfied) and to queue up a suspend
 117 request for the device in that case.
 118
 119 The helper functions provided by the PM core, described in Section 4, guarantee
 120 that the following constraints are met with respect to the bus type's run-time
 121 PM callbacks:
 122
 123 (1) The callbacks are mutually exclusive (e.g. it is forbidden to execute
 124     ->runtime_suspend() in parallel with ->runtime_resume() or with another
 125     instance of ->runtime_suspend() for the same device) with the exception that
 126     ->runtime_suspend() or ->runtime_resume() can be executed in parallel with
 127     ->runtime_idle() (although ->runtime_idle() will not be started while any
 128     of the other callbacks is being executed for the same device).
 129
 130 (2) ->runtime_idle() and ->runtime_suspend() can only be executed for 'active'
 131     devices (i.e. the PM core will only execute ->runtime_idle() or
 132     ->runtime_suspend() for the devices the run-time PM status of which is
 133     'active').
 134
 135 (3) ->runtime_idle() and ->runtime_suspend() can only be executed for a device
 136     the usage counter of which is equal to zero _and_ either the counter of
 137     'active' children of which is equal to zero, or the 'power.ignore_children'
 138     flag of which is set.
 139
 140 (4) ->runtime_resume() can only be executed for 'suspended' devices  (i.e. the
 141     PM core will only execute ->runtime_resume() for the devices the run-time
 142     PM status of which is 'suspended').
 143
 144 Additionally, the helper functions provided by the PM core obey the following
 145 rules:
 146
 147   * If ->runtime_suspend() is about to be executed or there's a pending request
 148     to execute it, ->runtime_idle() will not be executed for the same device.
 149
 150   * A request to execute or to schedule the execution of ->runtime_suspend()
 151     will cancel any pending requests to execute ->runtime_idle() for the same
 152     device.
 153
 154   * If ->runtime_resume() is about to be executed or there's a pending request
 155     to execute it, the other callbacks will not be executed for the same device.
 156
 157   * A request to execute ->runtime_resume() will cancel any pending or
 158     scheduled requests to execute the other callbacks for the same device.
 159
 160 3. Run-time PM Device Fields
 161
 162 The following device run-time PM fields are present in 'struct dev_pm_info', as
 163 defined in include/linux/pm.h:
 164
 165   struct timer_list suspend_timer;
 166     - timer used for scheduling (delayed) suspend request
 167
 168   unsigned long timer_expires;
 169     - timer expiration time, in jiffies (if this is different from zero, the
 170       timer is running and will expire at that time, otherwise the timer is not
 171       running)
 172
 173   struct work_struct work;
 174     - work structure used for queuing up requests (i.e. work items in pm_wq)
 175
 176   wait_queue_head_t wait_queue;
 177     - wait queue used if any of the helper functions needs to wait for another
 178       one to complete
 179
 180   spinlock_t lock;
 181     - lock used for synchronisation
 182
 183   atomic_t usage_count;
 184     - the usage counter of the device
 185
 186   atomic_t child_count;
 187     - the count of 'active' children of the device
 188
 189   unsigned int ignore_children;
 190     - if set, the value of child_count is ignored (but still updated)
 191
 192   unsigned int disable_depth;
 193     - used for disabling the helper funcions (they work normally if this is
 194       equal to zero); the initial value of it is 1 (i.e. run-time PM is
 195       initially disabled for all devices)
 196
 197   unsigned int runtime_error;
 198     - if set, there was a fatal error (one of the callbacks returned error code
 199       as described in Section 2), so the helper funtions will not work until
 200       this flag is cleared; this is the error code returned by the failing
 201       callback
 202
 203   unsigned int idle_notification;
 204     - if set, ->runtime_idle() is being executed
 205
 206   unsigned int request_pending;
 207     - if set, there's a pending request (i.e. a work item queued up into pm_wq)
 208
 209   enum rpm_request request;
 210     - type of request that's pending (valid if request_pending is set)
 211
 212   unsigned int deferred_resume;
 213     - set if ->runtime_resume() is about to be run while ->runtime_suspend() is
 214       being executed for that device and it is not practical to wait for the
 215       suspend to complete; means "start a resume as soon as you've suspended"
 216
 217   enum rpm_status runtime_status;
 218     - the run-time PM status of the device; this field's initial value is
 219       RPM_SUSPENDED, which means that each device is initially regarded by the
 220       PM core as 'suspended', regardless of its real hardware status
 221
 222 All of the above fields are members of the 'power' member of 'struct device'.
 223
 224 4. Run-time PM Device Helper Functions
 225
 226 The following run-time PM helper functions are defined in
 227 drivers/base/power/runtime.c and include/linux/pm_runtime.h:
 228
 229   void pm_runtime_init(struct device *dev);
 230     - initialize the device run-time PM fields in 'struct dev_pm_info'
 231
 232   void pm_runtime_remove(struct device *dev);
 233     - make sure that the run-time PM of the device will be disabled after
 234       removing the device from device hierarchy
 235
 236   int pm_runtime_idle(struct device *dev);
 237     - execute ->runtime_idle() for the device's bus type; returns 0 on success
 238       or error code on failure, where -EINPROGRESS means that ->runtime_idle()
 239       is already being executed
 240
 241   int pm_runtime_suspend(struct device *dev);
 242     - execute ->runtime_suspend() for the device's bus type; returns 0 on
 243       success, 1 if the device's run-time PM status was already 'suspended', or
 244       error code on failure, where -EAGAIN or -EBUSY means it is safe to attempt
 245       to suspend the device again in future
 246
 247   int pm_runtime_resume(struct device *dev);
 248     - execute ->runtime_resume() for the device's bus type; returns 0 on
 249       success, 1 if the device's run-time PM status was already 'active' or
 250       error code on failure, where -EAGAIN means it may be safe to attempt to
 251       resume the device again in future, but 'power.runtime_error' should be
 252       checked additionally
 253
 254   int pm_request_idle(struct device *dev);
 255     - submit a request to execute ->runtime_idle() for the device's bus type
 256       (the request is represented by a work item in pm_wq); returns 0 on success
 257       or error code if the request has not been queued up
 258
 259   int pm_schedule_suspend(struct device *dev, unsigned int delay);
 260     - schedule the execution of ->runtime_suspend() for the device's bus type
 261       in future, where 'delay' is the time to wait before queuing up a suspend
 262       work item in pm_wq, in milliseconds (if 'delay' is zero, the work item is
 263       queued up immediately); returns 0 on success, 1 if the device's PM
 264       run-time status was already 'suspended', or error code if the request
 265       hasn't been scheduled (or queued up if 'delay' is 0); if the execution of
 266       ->runtime_suspend() is already scheduled and not yet expired, the new
 267       value of 'delay' will be used as the time to wait
 268
 269   int pm_request_resume(struct device *dev);
 270     - submit a request to execute ->runtime_resume() for the device's bus type
 271       (the request is represented by a work item in pm_wq); returns 0 on
 272       success, 1 if the device's run-time PM status was already 'active', or
 273       error code if the request hasn't been queued up
 274
 275   void pm_runtime_get_noresume(struct device *dev);
 276     - increment the device's usage counter
 277
 278   int pm_runtime_get(struct device *dev);
 279     - increment the device's usage counter, run pm_request_resume(dev) and
 280       return its result
 281
 282   int pm_runtime_get_sync(struct device *dev);
 283     - increment the device's usage counter, run pm_runtime_resume(dev) and
 284       return its result
 285
 286   void pm_runtime_put_noidle(struct device *dev);
 287     - decrement the device's usage counter
 288
 289   int pm_runtime_put(struct device *dev);
 290     - decrement the device's usage counter, run pm_request_idle(dev) and return
 291       its result
 292
 293   int pm_runtime_put_sync(struct device *dev);
 294     - decrement the device's usage counter, run pm_runtime_idle(dev) and return
 295       its result
 296
 297   void pm_runtime_enable(struct device *dev);
 298     - enable the run-time PM helper functions to run the device bus type's
 299       run-time PM callbacks described in Section 2
 300
 301   int pm_runtime_disable(struct device *dev);
 302     - prevent the run-time PM helper functions from running the device bus
 303       type's run-time PM callbacks, make sure that all of the pending run-time
 304       PM operations on the device are either completed or canceled; returns
 305       1 if there was a resume request pending and it was necessary to execute
 306       ->runtime_resume() for the device's bus type to satisfy that request,
 307       otherwise 0 is returned
 308
 309   void pm_suspend_ignore_children(struct device *dev, bool enable);
 310     - set/unset the power.ignore_children flag of the device
 311
 312   int pm_runtime_set_active(struct device *dev);
 313     - clear the device's 'power.runtime_error' flag, set the device's run-time
 314       PM status to 'active' and update its parent's counter of 'active'
 315       children as appropriate (it is only valid to use this function if
 316       'power.runtime_error' is set or 'power.disable_depth' is greater than
 317       zero); it will fail and return error code if the device has a parent
 318       which is not active and the 'power.ignore_children' flag of which is unset
 319
 320   void pm_runtime_set_suspended(struct device *dev);
 321     - clear the device's 'power.runtime_error' flag, set the device's run-time
 322       PM status to 'suspended' and update its parent's counter of 'active'
 323       children as appropriate (it is only valid to use this function if
 324       'power.runtime_error' is set or 'power.disable_depth' is greater than
 325       zero)
 326
 327 It is safe to execute the following helper functions from interrupt context:
 328
 329 pm_request_idle()
 330 pm_schedule_suspend()
 331 pm_request_resume()
 332 pm_runtime_get_noresume()
 333 pm_runtime_get()
 334 pm_runtime_put_noidle()
 335 pm_runtime_put()
 336 pm_suspend_ignore_children()
 337 pm_runtime_set_active()
 338 pm_runtime_set_suspended()
 339 pm_runtime_enable()
 340
 341 5. Run-time PM Initialization, Device Probing and Removal
 342
 343 Initially, the run-time PM is disabled for all devices, which means that the
 344 majority of the run-time PM helper funtions described in Section 4 will return
 345 -EAGAIN until pm_runtime_enable() is called for the device.
 346
 347 In addition to that, the initial run-time PM status of all devices is
 348 'suspended', but it need not reflect the actual physical state of the device.
 349 Thus, if the device is initially active (i.e. it is able to process I/O), its
 350 run-time PM status must be changed to 'active', with the help of
 351 pm_runtime_set_active(), before pm_runtime_enable() is called for the device.
 352
 353 However, if the device has a parent and the parent's run-time PM is enabled,
 354 calling pm_runtime_set_active() for the device will affect the parent, unless
 355 the parent's 'power.ignore_children' flag is set.  Namely, in that case the
 356 parent won't be able to suspend at run time, using the PM core's helper
 357 functions, as long as the child's status is 'active', even if the child's
 358 run-time PM is still disabled (i.e. pm_runtime_enable() hasn't been called for
 359 the child yet or pm_runtime_disable() has been called for it).  For this reason,
 360 once pm_runtime_set_active() has been called for the device, pm_runtime_enable()
 361 should be called for it too as soon as reasonably possible or its run-time PM
 362 status should be changed back to 'suspended' with the help of
 363 pm_runtime_set_suspended().
 364
 365 If the default initial run-time PM status of the device (i.e. 'suspended')
 366 reflects the actual state of the device, its bus type's or its driver's
 367 ->probe() callback will likely need to wake it up using one of the PM core's
 368 helper functions described in Section 4.  In that case, pm_runtime_resume()
 369 should be used.  Of course, for this purpose the device's run-time PM has to be
 370 enabled earlier by calling pm_runtime_enable().
 371
 372 If the device bus type's or driver's ->probe() or ->remove() callback runs
 373 pm_runtime_suspend() or pm_runtime_idle() or their asynchronous counterparts,
 374 they will fail returning -EAGAIN, because the device's usage counter is
 375 incremented by the core before executing ->probe() and ->remove().  Still, it
 376 may be desirable to suspend the device as soon as ->probe() or ->remove() has
 377 finished, so the PM core uses pm_runtime_idle_sync() to invoke the device bus
 378 type's ->runtime_idle() callback at that time.