sys_pm_ctrl: fragility in managing state control

[pabigot]

subsys/power supports the option (via CONFIG_SYS_PM_STATE_LOCK) to enable and disable specific power states. State is managed with an array of atomic_t instances.

A request to disable a state corresponds to an atomic increment.

A request to enable a state corresponds to an atomic decrement.

The state is considered enabled when the counter is zero, which is the initial condition.

This approach results in some unusual behavior:

• sys_pm_ctrl_enable_state(S) when S is enabled decrements the counter to -1 thereby disabling the state.

• sys_pm_ctrl_disable_state(S) when S is disabled means that a subsequent sys_pm_ctrl_enable_state(S) may not enable the state.

The code has asserts to detect for underflow and overflow but since assertions are not generally enabled "misuse" is unlikely to be found.

It may be that this behavior is intended. If so it needs to be documented.

[github-actions]

This issue has been marked as stale because it has been open (more than) 60 days with no activity. Remove the stale label or add a comment saying that you would like to have the label removed otherwise this issue will automatically be closed in 14 days. Note, that you can always re-open a closed issue at any time.