Re: [PATCH] clk: fix clk_mux_val_to_index() error value

[Jerome Brunet]

On Tue, 2018-12-04 at 10:43 -0800, Stephen Boyd wrote:
> Quoting Jerome Brunet (2018-12-04 08:34:03)
> > clk_mux_val_to_index() is meant to be used by .get_parent(), which
> > returns a u8, so when the value provided does not map to any valid index,
> > it is not a good idea to return a negative error value.
> > 
> > Instead, return num_parents which we know is an invalid index and let
> > CCF deal with it.
> > 
> > Fixes: 77deb66d262f ("clk: mux: add helper function for index/value
> > translation")
> > Signed-off-by: Jerome Brunet <jbrunet@xxxxxxxxxxxx>
> > ---
> 
> Thanks!
> 
> > diff --git a/include/linux/clk-provider.h b/include/linux/clk-provider.h
> > index 60c51871b04b..fc20886ef069 100644
> > --- a/include/linux/clk-provider.h
> > +++ b/include/linux/clk-provider.h
> > @@ -550,8 +550,8 @@ struct clk_hw *clk_hw_register_mux_table(struct device
> > *dev, const char *name,
> >                 void __iomem *reg, u8 shift, u32 mask,
> >                 u8 clk_mux_flags, u32 *table, spinlock_t *lock);
> >  
> > -int clk_mux_val_to_index(struct clk_hw *hw, u32 *table, unsigned int
> > flags,
> > -                        unsigned int val);
> > +u8 clk_mux_val_to_index(struct clk_hw *hw, u32 *table, unsigned int
> > flags,
> 
> I wonder if we should just make this unsigned int? Does it hurt at all
> to have it be a wider type even though it doesn't match the CCF decision
> to make this a u8 for the parent index number space?
> 

I also wondered about this but since the target is get_parent(), I just
aligned the two.

In the end, I don't really care, as you prefer. Just let me know if you would
like a v2 with this change