Re: [PATCH v4 1/4] pwm: pca9685: Switch to atomic API

[Sven Van Asbroeck]

On Tue, Dec 8, 2020 at 11:57 AM Thierry Reding <thierry.reding@xxxxxxxxx> wrote:
>
> Is this really that complicated? I sounds to me like the only thing that
> you need is to have some sort of usage count for the prescaler. Whenever
> you want to use the prescaler you check that usage count. If it is zero,
> then you can just set it to whatever you need. If it isn't zero, that
> means somebody else is already using it and you can't change it, which
> means you have to check if you're trying to request the value that's
> already set. If so, you can succeed, but otherwise you'll have to fail.

+1
I think your suggestion is an elegant solution to get the required behaviour.

One possible complication is synchronization. The sysfs interface has a lock
protecting against concurrent pwm_apply() calls. But the in-kernel
API (e.g. pwm_apply_state()) doesn't seem to. This is not normally a problem
when pwm bits are strictly separated. But in this case we have shared state
(prescale value and use count), so we probably need to protect pwm_apply()
with a mutex?

Not sure if it is currently possible *in practice* for two regulator consumer
drivers to call pwm_apply() from different threads. But Linux is slowly moving
towards asynchronous probing.

Uwe and Thierry, what is your opinion? Do you think we need to worry about
synchronization?