Re: lm90 driver no longer working on PCs in 3.13

[Guenter Roeck <linux@roeck-us.net>]

On 01/26/2014 12:52 PM, Mark Brown wrote:
> On Sun, Jan 26, 2014 at 12:44:38PM -0800, Guenter Roeck wrote:
>> On 01/26/2014 12:13 PM, Jean Delvare wrote:
>
>>> Me, I really don't know. I seem to remember I tested Wei's patch set on
>>> an emulated ADM1032 chip and it was working fine. So maybe it depends
>>> on the kernel configuration, or something changed on the regulator side
>>> meanwhile.
>
>> The regulator code changed with 3.13; the dummy regulator no longer exists,
>> and the functionality it provided is supposed to be handled automatically.
>> But that only really works on devicetree based systems and otherwise returns
>> -EPROBE_DEFER as mentioned above.
>
> CONFIG_REGULATOR_DUMMY should never have been used in production, it was
> a debug tool to help bringup but it broke things as often as it fixed
> them particularly with init ordering which is why it generated a warning
> when it was used.
>
> The dummy driver is still there, if you're doing bringup you can hack it
> in still or if you genuniely used it then specify that full constraints
> are provided like the changelog says (and as I've previously said).
>
>> Another possible fix would be to have the regulator core return -ENODEV
>> instead of -EPROBE_DEFER on non-dt systems. No idea if this would be acceptable
>> or even feasible.
>
> No, this would introduce breakage due to init ordering.
>

You have a solution for that in dt configurations. I don't think you have one for
non-dt systems - you simply assume that all regulators are there. For dt, you even
have a constraint to tell the kernel if regulator configurations are fully specified,
and you automatically return success if not and if a regulator does not exist.
So you know that there is a problem. For non-DT configurations you simply assume and
expect that regulators are all declared. I don't think that is a feasible approach
for non-DT systems.

Guenter