On 02.12.22 09:50, Marco Felsch wrote: > On 22-12-01, Ahmad Fatoum wrote: >> regulator_get calls either of_regulator_get or dev_regulator_get, both >> of which accept a supply parameter. Only dev_regulator_get gracefully >> handles a NULL supply by considering all registered regulators. > > Wouldn't it be better to align of_regulator_get behaviour with > dev_regulator_get?
Nope, not specifying a supply id is legacy behavior. Consumers should be expected to know the name of the supply they want to control. In Linux, regulator_get(dev, NULL) is already an error. > >> of_regulator_get on the other hand, will complain at debug level, >> before returning NULL: >> >> uart-pl011 [email protected]: No <NULL>-supply node found, >> using dummy regulator >> >> Avoid this message by skipping of_regulator_get if no supply was found >> and directly call dev_regulator_get. >> >> Signed-off-by: Ahmad Fatoum <[email protected]> >> --- >> drivers/regulator/core.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c >> index 4876f0f44bdd..d25aba38c3e2 100644 >> --- a/drivers/regulator/core.c >> +++ b/drivers/regulator/core.c >> @@ -388,7 +388,7 @@ struct regulator *regulator_get(struct device_d *dev, >> const char *supply) >> struct regulator *r; >> int ret; >> >> - if (dev->device_node) { >> + if (dev->device_node && supply) { >> ri = of_regulator_get(dev, supply); >> if (IS_ERR(ri)) >> return ERR_CAST(ri); > > Nevertheless the change is correct, so: > > Reviewed-by: Marco Felsch <[email protected]> > > Regards, > Marco >> -- >> 2.30.2 >> >> >> > -- Pengutronix e.K. | | Steuerwalder Str. 21 | http://www.pengutronix.de/ | 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
