Re: [PATCH] regulator: core: Keep boot_on regulators powered during init

[Ulf Hansson <ulf.hansson@stericsson.com>]

Hi Mark,

Thanks for a quick reply.

On 04/23/2012 12:18 PM, Mark Brown wrote:
> On Mon, Apr 23, 2012 at 11:37:53AM +0200, Ulf Hansson wrote:
>
>> Regulators which has boot_on constraints set, will now remain
>> powered after regulator_init_complete is done.
>
> This would be a bug.  All boot_on means is that the regulator was turned
> on during boot, the regulator is free to vary after that.

The idea is to prevent the late_init_call "regulator_init_complete" from 
disabling a regulator that "recently" were enabled due to it's boot_on 
constraints.

>
>> In this case we leave the enable->disable operation to be
>> handled by the regulator consumer instead.
>
> Which would be a bug if the consumer wasn't the thing that took the
> reference to the regulator in the first place.  Remember, regulators can
> be shared so the consumer can't disable a regulator it didn't enable in
> the first place (unless it used regulator_get_exclusive() but the use
> cases for that are a little suspicious so it'd be worth taking a careful
> look before using it).  A consumer can't tell if the regulator was left
> enabled by the firmware on boot or if it has been enabled by another
> consumer.

I realize that using boot_on, which has been around for quite some time 
could have problems.  If not using the existing boot_on constraint, do 
you have an idea of how to accomplish what I want? Should I invent a new 
constraint option to be used in regulator_init_complete!?

Kind regards
Ulf Hansson