3.1 Guidelines for wireless device drivers
------------------------------------------
+(in this text, rfkill->foo means the foo field of struct rfkill).
+
1. Each independent transmitter in a wireless device (usually there is only one
transmitter per device) should have a SINGLE rfkill class attached to it.
when possible) the overall transmitter rfkill state, not of a particular rfkill
line.
+5. The wireless device driver MUST NOT leave the transmitter enabled during
+suspend and hibernation unless:
+
+ 5.1. The transmitter has to be enabled for some sort of functionality
+ like wake-on-wireless-packet or autonomous packed forwarding in a mesh
+ network, and that functionality is enabled for this suspend/hibernation
+ cycle.
+
+AND
+
+ 5.2. The device was not on a user-requested BLOCKED state before
+ the suspend (i.e. the driver must NOT unblock a device, not even
+ to support wake-on-wireless-packet or remain in the mesh).
+
+In other words, there is absolutely no allowed scenario where a driver can
+automatically take action to unblock a rfkill controller (obviously, this deals
+with scenarios where soft-blocking or both soft and hard blocking is happening.
+Scenarios where hardware rfkill lines are the only ones blocking the
+transmitter are outside of this rule, since the wireless device driver does not
+control its input hardware rfkill lines in the first place).
+
+6. During resume, rfkill will try to restore its previous state.
+
+7. After a rfkill class is suspended, it will *not* call rfkill->toggle_radio
+until it is resumed.
+
+
Example of a WLAN wireless driver connected to the rfkill subsystem:
--------------------------------------------------------------------
rfkill input line is active. Only if none of the rfkill input lines are
active, will it return RFKILL_STATE_UNBLOCKED.
-If it doesn't implement the get_state() hook, it must make sure that its calls
-to rfkill_force_state() are enough to keep the status always up-to-date, and it
-must do a rfkill_force_state() on resume from sleep.
+Since the device has a hardware rfkill line, it IS subject to state changes
+external to rfkill. Therefore, the driver must make sure that it calls
+rfkill_force_state() to keep the status always up-to-date, and it must do a
+rfkill_force_state() on resume from sleep.
Every time the driver gets a notification from the card that one of its rfkill
lines changed state (polling might be needed on badly designed cards that don't
about its current state).
The rfkill class will call the get_state hook of a device every time it needs
-to know the *real* current state of the hardware. This can happen often.
+to know the *real* current state of the hardware. This can happen often, but
+it does not do any polling, so it is not enough on hardware that is subject
+to state changes outside of the rfkill subsystem.
+
+Therefore, calling rfkill_force_state() when a state change happens is
+mandatory when the device has a hardware rfkill line, or when something else
+like the firmware could cause its state to be changed without going through the
+rfkill class.
Some hardware provides events when its status changes. In these cases, it is
best for the driver to not provide a get_state hook, and instead register the
rfkill class *already* with the correct status, and keep it updated using
rfkill_force_state() when it gets an event from the hardware.
+rfkill_force_state() must be used on the device resume handlers to update the
+rfkill status, should there be any chance of the device status changing during
+the sleep.
+
There is no provision for a statically-allocated rfkill struct. You must
use rfkill_allocate() to allocate one.