Re: [PATCH 0/2] cpufreq/opp: rework regulator initialization

[Viresh Kumar]

On 08-02-19, 11:18, Rafael J. Wysocki wrote:
> Obviously, all I2C transfers need to take place either before
> suspending the I2C controller or after resuming it.

Right. But all I am saying is that it is the responsibility of the
cpufreq core/driver to make sure we call ->init() only when the
resources are available to be used. For example during driver load, we
defer probe if resources aren't available and that makes sure ->init()
gets called when we can actually change the frequency.

Shouldn't the same be true during resume? That is make sure all
resources are in place before ->init() gets called ?

-- 
viresh