Hi Quentin, On Feb 4, 2025 at 09:16:29, Quentin Schulz <quentin.sch...@cherry.de> wrote:
> Hi Justin, > > On 1/28/25 10:37 PM, Justin Klaassen wrote: > > [You don't often get email from jus...@tidylabs.net. Learn why this is > important at https://aka.ms/LearnAboutSenderIdentification ] > > > Log the value of the regulators during initialization of the IO-domain > > driver to aid in debugging GPIO voltage configuration problems. > > > Signed-off-by: Justin Klaassen <jus...@tidylabs.net> > > --- > > > Changes in v2: > > - Made added IO-domain driver logs verbose and consistent > > > drivers/misc/rockchip-io-domain.c | 7 ++++++- > > 1 file changed, 6 insertions(+), 1 deletion(-) > > > diff --git a/drivers/misc/rockchip-io-domain.c > b/drivers/misc/rockchip-io-domain.c > > index 025b6049a9f..fdefc9c5fe3 100644 > > --- a/drivers/misc/rockchip-io-domain.c > > +++ b/drivers/misc/rockchip-io-domain.c > > @@ -5,6 +5,8 @@ > > * Ported from linux drivers/soc/rockchip/io-domain.c > > */ > > > +#define LOG_CATEGORY UCLASS_NOP > > + > > #include <dm.h> > > #include <dm/device_compat.h> > > #include <regmap.h> > > @@ -344,8 +346,10 @@ static int rockchip_iodomain_probe(struct udevice > *dev) > > continue; > > > ret = device_get_supply_regulator(dev, supply_name, ®); > > - if (ret) > > + if (ret) { > > + log_debug("%s: Regulator not found\n", > supply_name); > > > Can you use dev_dbg() instead? > Done. > This would match what we use for other messages in the probe function? > > Specifically, it could add the device name in front, which would be > useful when debugging to highlight the IO domain device is the one > requesting this regulator that isn't found. > > continue; > > + } > > > ret = regulator_autoset(reg); > > if (ret && ret != -EALREADY && ret != -EMEDIUMTYPE && > > @@ -353,6 +357,7 @@ static int rockchip_iodomain_probe(struct udevice *dev) > > continue; > > > uV = regulator_get_value(reg); > > + log_debug("%s: Regulator %s at %d uV\n", supply_name, > reg->name, uV); > > > Ditto. > Done. > Not sure the reg->name brings a lot since this information could be > derived from supply_name if I'm not mistaken, but I'm not too bothered > by it 🙂 > I found reg->name to be helpful when debugging as it made it slightly easier to find the correct regulator without having to go through a few levels of redirection in the device tree. > > Cheers, > Quentin > > Thanks, Justin
