On Thu, 2016-10-20 at 14:13 +0300, Ville Syrjälä wrote: > On Thu, Oct 20, 2016 at 12:36:13PM +0300, Andy Shevchenko wrote: > > The Crystal Cove GPIO driver has its explicit name, i.e. > > crystal_cove_gpio. > > > > Use it explicitly to be in align with the rest of code, pretty > > grepping, and > > prevent potential breakage if GPIO driver file name ever changed. > > > > Cc: Linus Walleij <[email protected]> > > Cc: Ville Syrjälä <[email protected]> > > Signed-off-by: Andy Shevchenko <[email protected]> > > --- > > drivers/mfd/intel_soc_pmic_core.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/drivers/mfd/intel_soc_pmic_core.c > > b/drivers/mfd/intel_soc_pmic_core.c > > index 12d6ebb4..7fd8e87 100644 > > --- a/drivers/mfd/intel_soc_pmic_core.c > > +++ b/drivers/mfd/intel_soc_pmic_core.c > > @@ -34,7 +34,7 @@ static struct gpiod_lookup_table panel_gpio_table > > = { > > .dev_id = "0000:00:02.0", > > .table = { > > /* Panel EN/DISABLE */ > > - GPIO_LOOKUP("gpio_crystalcove", 94, "panel", > > GPIO_ACTIVE_HIGH), > > + GPIO_LOOKUP("crystal_cove_gpio", 94, "panel", > > GPIO_ACTIVE_HIGH), > > This looks like it would break things.
Yeah. Good catch. > The gpio_chip->label which is > used for the lookup is based on KBUILD_MODNAME and the moduled is > called gpio-crystalcove. ...which I think is wrong. Only few drivers are binding to file / module name. Perhaps it should be fixed all together. Naming a bit chaotic in drivers/gpio/*::chip.label, though majority of them are using dev_name() of actual device. Linus, what is your opinion? -- Andy Shevchenko <[email protected]> Intel Finland Oy
