On Fri, May 25, 2018 at 11:14:23AM +0100, Mark Brown wrote:

On Fri, May 25, 2018 at 10:22:53AM +0300, Matti Vaittinen wrote:

quoted

On Thu, May 24, 2018 at 05:51:27PM +0000, Vaittinen, Matti wrote:

quoted

quoted

quoted

devm_regulator_regster()

quoted

quoted

Makes sense. Thanks

quoted

I was going to do 

quoted

-       pmic = kzalloc(sizeof(struct bd71837_pmic), GFP_KERNEL);
+       pmic = devm_kzalloc(&pdev->dev, sizeof(struct bd71837_pmic),
+                           GFP_KERNEL);

quoted

and

quoted

-               rdev = regulator_register(desc, &config);
+               rdev = devm_regulator_register(&pdev->dev, desc, &config);

but is there now a race regarding freeing the pmic structure and
unregistering the regulator?

Why?  devm_ stuff gets unwound in the opposite order to the order in
which it was allocated.

Allright. Then there's no problems. I'll do the change.