Re: [RFC 1/2] PM / suspend: Add platform_suspend_target_state()

[Rafael J. Wysocki]

On Thursday, July 06, 2017 05:17:50 AM Pavel Machek wrote:
> On Sun 2017-07-16 01:29:57, Rafael J. Wysocki wrote:
> > On Saturday, July 15, 2017 06:46:26 PM Pavel Machek wrote:
> > > Hi!
> > > 
> > > > > > I had an idea of using an enum type encompassing all of the power states
> > > > > > defined for various platforms and serving both as a registry (to ensure the
> > > > > > uniqueness of the values assigned to the states) and a common ground
> > > > > > between platforms and drivers.
> > > > > > 
> > > > > > Something like:
> > > > > > 
> > > > > > enum platform_target_state {
> > > > > > 	PLATFORM_STATE_UNKNOWN = -1,
> > > > > > 	PLATFORM_STATE_WORKING = 0,
> > > > > > 	PLATFORM_STATE_ACPI_S1,
> > > > > > 	PLATFORM_STATE_ACPI_S2,
> > > > > > 	PLATFORM_STATE_ACPI_S3,
> > > > > > 	PLATFORM_STATE_MY_BOARD_1_GATE_CLOCKS,
> > > > > > 	PLATFORM_STATE_MY_BOARD_1_GATE_POWER,
> > > > > > 	PLATFORM_STATE_ANOTHER_BOARD_DO_CRAZY_STUFF,
> > > > > > 	...
> > > > > > };
> > > > > > 
> > > > > > and define ->target_state to return a value of this type.
> > > > > > 
> > > > > > Then, if a driver sees one of these and recognizes that value, it should
> > > > > > know exactly what to do.
> > > > > 
> > > > > Remind me why this is good idea?
> > > > 
> > > > Because there are drivers that need to do specific things during
> > > > suspend on a specific board when it goes into a specific state as a
> > > > whole.
> > > 
> > > We have seen driver that cares about voltage to his device being
> > > lost. That's reasonable.
> > > 
> > > Inquiring what the platform target state is... is not.
> > 
> > So why exactly isn't it reasonable?
> > 
> > Please use technical arguments.  Saying that something is wrong without
> > explaining the problem you see with it isn't particulatly useful in technical
> > discussions.
> 
> Deep in your heart, you should know that having enum listing all the platforms linux
> runs on is a very bad idea.

Even so, if I'm unable to explain to people why this is a bad idea in technical
terms, that doesn't mean too much.

I actually noticed an issue with the approach that I missed before, see my last
reply to Florian.

> Anyway, there are better solutions, regulator framework already knows if given rail
> will be powered off or not, and their driver already knows if they are going
> suspend/standby. They just need to use existing interfaces.

So they need to know what has been passed to suspend_devices_and_enter()
anyway and currently there's no interface for that.  That actually is the source
of the whole issue.

Thanks,
Rafael