Currently power management subsystem uses a global variable to hold
power state. On multicore environment this is a problem and we have to
have this information per core.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move all PM device runtime API calls from pm_device* to the
pm_device_runtime* namespace.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
This patch refactors the runtime API to make it more clear and simple.
Relevant changes:
- API uses the action callback in a direct manner, avoiding unnecessary
overhead.
- API documentation has been improved to include detailed return error
codes.
- pm_runtime_disable() is now synchronous (to simplify possible error
paths) and returns error in case it fails. It is also safe to disable
in pre-kernel now.
- pm_runtime_put(_async)() will return -EALREADY if called with usage
count at zero (result of an unbalanced get/put call sequence)
- A transitional state has been added back. This makes code more
readable, and avoids using atomics (not required).
TODO:
- Solve in a better manner the asynchronous suspend error path (now
"solved" using asserts).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
After the removal of pm_device_get_async, the pm_device_wait API has
become redundant. Its usage after pm_device_put_async should not be
considered a valid usecase, since after that call what will happen is a
pm_device_get (which is blocking).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
As of today there is no clear usage of asynchronous gets, since in
general, a resume operation should be synchronous (we are about to use
the device immediately after resuming it). Removing this API simplifies
the runtime implementation in a significant way (refer to future
commits).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Move functions around so it is not necessary to keep a header
that with functions declaration that is just used in one single
place.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
List of improvements:
- The PM logging module was only available if CONFIG_PM=y, however, it
was also used by Device PM (which can be selected without PM). A new
logging module has been created for Device PM.
- Log level is passed to LOG_MODULE_(DECLARE|REGISTER)
- Logger name has been adjusted to `pm` (was `power`)
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
POLL is not a dependency of runtime device PM since it now uses
conditional variables to notify waiting threads.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
SYS_POWER_MANAGEMENT and DEVICE_POWER_MANAGEMENT were deprecated in
2.5.0, remove them now.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
There is not need to handle all power states in switch state after
remove lower power and off states for devices.
Also, it incorporates the fix done in pull/38873.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
As discussed in issues/38619, this commit aims to simplify device
power management. It was removed PM_DEVICE_SATE_LOW_POWER.
The power subsystem now doesn't need to figure out which device state to
use of a given system power state. It just suspend and resume devices.
Devices now just need to respond to ACTIVE and SUSPEND and OFF actions
and they are free to use any particular substate they have when the
subsystem asks to suspend. They also don't need to worry about states
transitions (unless they have multiple substates) because the system
will just request them to suspend if they are active and vice-versa.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
rebase
Add a function that can be used by device drivers to know
what will be next power state used by the SoC.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Uses the stats subsys to provide simple but useful debugging stats for
power management state changes and timing.
Removes the no longer needed PM_DEBUG config option
Replaces the use of PM_DEBUG for a test clock output pin for mec1501 and
adds in its place an SoC Kconfig option to enable it.
Adds a STATS_SET macro for assigning a value to a stat group field
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
- Rename to "action" to make its purpose more clear
- Use the _cb suffix to align with naming used for callbacks in other
areas.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Accessing members from pm_device improves code readability, since it
removes dev-> from most accesses.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Some includes were already performed by the device(_runtime).h header/s,
others like were not necessary.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
If a driver doesn't support PM, as indicated by setting the
pm_control_fn parameter to NULL, no need to manage busy or wakeup state.
This also prepares the PM support for issue #39286, which will allocate
PM structures only for the devices that request it.
Signed-off-by: Keith Short <keithshort@google.com>
Devices need to be resumed in the reverse order they are suspended.
e.g: devA +---> devB ---> devD
|
+---> devC
They are initialized in the following order, devA -> devB -> devC ->
devD, and suspended starting from the end of the list, devD -> devC ->
devB -> devA. When they are suspended they are temporary put in a list
that is used later to resume them.
This list has to be iterated from the end to the beginning, otherwise a
device may be resumed before its parent.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Previously stats were kept in a single static but would be updated by an
idle thread per cpu core. Stats/debug info is now kept per cpu core.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
According to the state documentation, this state does not need to handle
devices:
> Runtime idle is a system sleep state in which all of the cores enter
deepest possible idle state and wait for interrupts, no requirements for
the devices, leaving them at the states where they are.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
In order to be able to make a "choice" from any other
Kconfig.defconfig (-ish) file, the choice needs to be named.
This commit names a few choices.
Signed-off-by: Emil Lindqvist <emil@lindq.gr>
Implement constraint API as weak symbols so applications or platform
can override them. Constraints are a high-level abstraction to inhibit
the power subsystem of using certain power states. Some platforms can
have their own way to set/release constraints in their drivers that
are not part of Zephyr's code base.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
PM_DEVICE_STATE_RESUMING and PM_DEVICE_STATE_SUSPENDING
are transitional states and are only used in device runtime. Remove it
and use device flag to keep track of a transition.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
The runtime API is referenced count and uses a uint32_t.
Avoid underflow when dealing with put requests.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Return 0 in cases where the request does not trigger a device state
change, only incremented or to decremented the reference count.
Fixes#37821
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Suspend devices when the system goes to STATE_SOFT_OFF.
This state was not triggering any device power management. This should
at least suspend devices as it is done for SUSPEND_TO_RAM and
SUSPEND_TO_DISK.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
_pm_devices, pm_suspend_devices, pm_low_power_devices and
pm_resume_devices are only used if CONFIG_PM_DEVICE is defined and not
CONFIG_PM.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Introduce a new API to allow devices capable of wake up the system
register themselves was wake up sources. This permits applications to
select the most appropriate way to wake up the system when it is
suspended.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Just using a simple atomic for flags instead of using an array.
While is neat using ATOMIC_DEFINE for future proof. The reality is
that it brings some problem for the wakeup source implementation
that needs to statically initialize it during the device definition.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Use the internal function z_device_get_all_static helper function
instead of using __device_start and __device_end directly. Some other
minor adjustments have been done (e.g. reduce *dev scope to the for
loop). An issue on the range of the for loop in _pm_devices has also
been fixed.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The shared _pm_devices function used should_suspend check function to
see if a device had to be suspended or not. Some of the logic inside
that function was redundant since the pm_device_state_set function
already performs similar checks, e.g. if the device is already at the
given state or the state transition is not supported it will return
error codes appropriately.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
A device that is turned off should not be suspended. A device that has
been turned off can only be resumed. This action is currently forbidden
by the "should_suspend" function in the device PM code.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Add support for FORCE_SUSPEND and LOW_POWER actions. Even though these
actions have no clear meaning, they are added for compatibility until
their associated states are discussed. Their usage in new code should be
discouraged until the associated states are clarified.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
If a device does not support, for example, the suspend action the
subsystem should just ignore the device.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Instead of passing target states, use actions for device PM control.
Actions represent better the meaning of the callback argument.
Furthermore, they are more future proof as they can be suitable for
other PM actions that have no direct mapping to a state. If we compare
with Linux, we could have a multi-stage suspend/resume. Such scenario
would not have a good mapping when using target states.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The verb tense for the suspended state was not consistent with other
states. The likely reason: state was being used as a command/action.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
If the device is already at the given state, do not call the device PM
control function. This makes sure that devices are only called to change
from one state to another.
Even though asynchronous device PM is completely broken, transitional
states are considered too.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
