Some devices, e.g. SoC level devices like I2C peripheral, can never be
powerd off as they are always energized. Such devices can only go from an
active state or to a low power state (suspended). Allow them to simply
return -ENOTSUP when called with TURN_ON (or TURN_OFF).
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Some events needs to be handled with a very low latency constraint.
If the system is in deep sleep, exit latency from this low level state
exceeds sometimes the maximum latency constraint of these events.
Before suspending the system, select which events is happening sooner,
kernel events or normal events.
CPU will be up just before the next event occurs taking into account the
exit latency of the current power state
Change also the policy event API to take as argument absolute time in HW
cycles instead of time in us
Signed-off-by: Riadh Ghaddab <rghaddab@baylibre.com>
Refactor pm_device_driver_init code to keep the normal execution path
inline and the early exit branches at a single indentation, this is
commonly done throughout the code base.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
PM_DEVICE_RUNTIME_EXCLUSIVE was deprecated and its behavior
is achived with PM_DEVICE_SYSTEM_MANAGED=n.
Fixes#76037
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move information about device power state constraints from device
to policy.
It slows down the constraints lookup since we now have to find the
constraints for a device in a global array, but it saves resources
because we don't need to add a reference to constraints in all devices
instances.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Add a symbol to enable device power state constraints this
saves resources when this feature is not needed.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Declare power state constraints for a device in devicetree.
It allows a map between device instances and power states that disable
their power. This information is used by a new API
(pm_policy_device_power_lock_put/get) that automically set/release
pm state constraints.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
PM_DEVICE is not attached to system managed device power management.
It is a very common use case targets with device runtime power
management that don't want system device power management enabled.
We introduce a new symbol (PM_DEVICE_SYSTEM_MANAGED) to explicit
control whether or not system device power management should be
globally enabled.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Remove device pm path when there is no is no power state in DT with
device pm enabled. This basically does the same thing that was done
by PM_DEVICE_RUNTIME_EXCLUSIVE.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Make it possible to disble device power management individually per
power state. This allows targets tuning which states should
(and which should not) trigger device power management.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
That is option has shown confusing on it is attempt to prevent
system pm doing device power management. Lets address this
problem properly.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
pm_system_resume() can be called directly from ISR and because of this
devices should be resumed before calling pm_state_exit_post_ops().
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
There is no need to this function be defined inside the kernel since
all places using it are protecting the call under ifdef PM guards.
This way we can also remove the ifdef condition inside the implementation.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
s2ram procedure used RAM magic word for marking suspend-to-RAM. This
method may not work in some cases, e.g. when global reset does not
reset RAM content. In that case resuming from s2ram is detected when
global reset occurred.
RAM magic word method is the default but with
CONFIG_PM_S2RAM_CUSTOM_MARKING a custom implementation can be provided.
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
We can't enable device runtime power management in a device that is
set busy since it may suspend this device.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Add support for a "pm" shell command to trigger suspend/resume as well
as runtime-get/put on devices. This is useful for testing during driver
development.
Signed-off-by: Fabio Baltieri <fabiobaltieri@google.com>
In the follow expression:
cyc_evt += UINT32_MAX + 1U
first it is evaluated (UINT32_MAX + 1U), since both types
interpreted as uint32_t, this operation causes an overflow resulting
in 0U.Then we have
cyc_evt = (uint64_t)cyc_evt + 0U
Fix it casting of the operands in the first operation to uint64_t.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
This API is not widely used and it is actually broken since device
runtime power management is not checking it when suspending and
resuming.
On top of that, this API is very close to pm_device_busy* API,
close enough to consolidate in only one API.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
To wait for the asynchronous suspending work item to complete, a
combination of semaphores and events is used. First, the semaphore is
released, then the events are cleared (through the boolean argument to
k_event_wait), then events are awaited.
However, if the event flag happens to be set by the work handler in the
short time between k_sem_give and k_event_wait, it is then cleared by
k_event_wait and k_event_wait blocks forever waiting for the event.
Make sure that we clear the event flag before releasing the semaphore.
Signed-off-by: Marco Widmer <marco.widmer@bytesatwork.ch>
When the policy returns NULL pm_system_suspend was assuming that
the current state in z_cpus_pm_state was ACTIVE, since that is the
state set after the core wakes and return to this function. The
problem is that in cases where the cpu does not preserve the context,
and returns to this function, z_cpus_pm_state has the value of the
last state used and the cpu use it again.
Fix it setting z_cpus_pm_state to ACTIVE every time the policy returns
NULL.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move the check of PM_DEVICE_FLAG_RUNTIME_ENABLED to the beginning of
the function. With this we avoid taking/release the semaphore and also
we no longer need check it in runtime_enable_sync(), because it was
already checked in pm_device_runtime_enable().
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move around SYS_PORT_FUNC_ENTER and change a early return
in a way that when this function is called it will generate
tracing for entering / leaving this function.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move the check of PM_DEVICE_FLAG_RUNTIME_ENABLED to the beginning of
the function. With this we avoid taking/release the semaphore and also
we no longer need check it in runtime_disable_sync(), because it was
already checked in pm_device_runtime_disable().
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Move around SYS_PORT_FUNC_ENTER and change a early return
in a way that when this function is called it will generate
tracing for entering / leaving this function.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
In many cases suspending or resuming of a device is limited to
just a few register writes. Current solution assumes that those
operations may be blocking, asynchronous and take a lot of time.
Due to this assumption runtime PM API cannot be effectively used
from the interrupt context. Zephyr has few driver APIs which
can be used from an interrupt context and now use of runtime PM
is limited in those cases.
Patch introduces a new type of PM device - synchronous PM. If
device is specified as capable of synchronous PM operations then
device runtime getting and putting is executed in the critical
section. In that case, runtime API can be used from an interrupt
context. Additionally, this approach reduces RAM needed for
PM device (104 -> 20 bytes of RAM on ARM Cortex-M).
Signed-off-by: Krzysztof Chruściński <krzysztof.chruscinski@nordicsemi.no>
If runtime_suspend() is called early during Zephyr initialisation,
while k_is_pre_kernel() returns 'true,' 'async' is set to 'false,' so
if 'async' is 'true,' Zephyr initialisation is definitely complete,
so there is no need to check k_is_pre_kernel() again.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
The asynchronous put is not checking if the device was successfully
suspended before suspending its domain and it is not checking if the
domain was claimed. This patch adds these two checks.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
pm_suspend_devices() could return an error. Set timeout using
sys_clock_set_timeout() to after this error is handled so that
we have the accurate power state when calling the timeout function.
This is useful in cases where we wish to compensate the
system timer for certain power modes.
Signed-off-by: Mahesh Mahadevan <mahesh.mahadevan@nxp.com>
Similar to what is done in pm_device_runtime_get,
we don't need to wait the device be suspended if
the work is still in the work queue. In this case
we just cancel the work and update the device state
to active.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Add a delay parameter to asynchronous device runtim put. This allows
to delay the put operation what is useful to avoid multiple states
transitions.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
We don't need to wait an async put to happen in case it has not
started yet. In this case we can simply cancelling the pending work
and change the internal state because the device is still active.
This way we avoid a suspend and resume cycle.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
Don't assert that `pm_device_runtime_get` is not running in an ISR
context, as `runtime_suspend` properly handles that condition.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Enable the automatic power domain management to release the domain as
many times as it was claimed, instead of only once.
This fixes the domain being permanently enabled if the supported device
is claimed more than once.
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
Add check for device busy when CONFIG_PM_NEED_ALL_DEVICES_IDLE is
set to y because one or more devices may still in busy and causes
problem when system go into low power in Intel ISH platform.
Signed-off-by: Leifu Zhao <leifu.zhao@intel.com>
This is a private kernel header with private kernel APIs, it should not
be exposed in the public zephyr include directory.
Once sample remains to be fixed (metairq_dispatch), which currently uses
private APIs from that header, it should not be the case.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
In some platforms it may be desirable to disable certain CPU power
states, for example, because they have extra requirements not available
on all boards/applications. Because `cpu-power-states` are defined at
SoC dts file levels, the only way to achieve that now was by re-defining
`cpu-power-states` property in e.g. a board file. With this patch, one
can now selectively set `status = "disabled";` to any power state and it
will be skipped by the PM subsystem.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
pm_device_runtime_get() uses a semaphore to protect resources.
pm_device_runtime_get() blocked forever until obtaining the lock, making
it unusable from contexts where blocking is not allowed, e.g. ISRs. With
this patch, we give the chance to use the function from an ISR by not
waiting on the lock and returning -EWOULDBLOCK instead. If device is
suspending (from a previous put_async() operation) the same value is
returned as we can't wait either.
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
Adds a helper function for initializing devices into the expected power
state, through the devices `pm_device_action_cb_t`. This eliminates code
duplication between the init functions and the PM callback.
The expected device states in order of priority are:
* No power applied to device, `OFF`
* `zephyr,pm-device-runtime-auto` enabled, `SUSPEND`
* Otherwise, `ACTIVE`
Signed-off-by: Jordan Yates <jordan.yates@data61.csiro.au>
