[U-Boot] [PATCH v2 01/50] dm: clk: Add support for decoding clocks from the device tree

Simon Glass sjg at chromium.org
Wed Jan 20 16:24:50 CET 2016


Hi Masahiro,

On 19 January 2016 at 22:38, Masahiro Yamada
<yamada.masahiro at socionext.com> wrote:
>
> Hi Simon,
>
>
> >>>
> >>> +/**
> >>> + * clk_get_by_index() - look up a clock referenced by a device
> >>> + *
> >>> + * Parse a device's 'clocks' list, returning information on the indexed clock,
> >>> + * ensuring that it is activated.
> >>> + *
> >>> + * @dev:       Device containing the clock reference
> >>> + * @index:     Clock index to return (0 = first)
> >>> + * @clk_devp:  Returns clock device
> >>> + * @return:    Peripheral ID for the device to control. This is the first
> >>> + *             argument after the clock node phandle. If there is no arguemnt,
> >>> + *             returns 0. Return -ve error code on any error
> >>> + */
> >>> +int clk_get_by_index(struct udevice *dev, int index, struct udevice **clk_devp);
> >>>  #endif /* _CLK_H_ */
> >>
> >>
> >> I want #ifdef in the header too, like mine
> >> http://patchwork.ozlabs.org/patch/566812/
> >
> > I am not keen on that idea since it clutters up header files and we'll
> > get a link error anyway if something is missing. Anyway, I've added
> > it.
>
>
>
> I am afraid there is misunderstanding here.
>
> Please see my patch carefully.
>
>
> What I mean is like this:
>
>
>   #if  ...
>       declaration of function prototype
>   #else
>       static inline empty function
>   #endif
>
>
> This is a common technique to avoid a link error.

Do you think this function will be called when device tree is not
enabled? I cannot see how it would have any meaning in that case. What
is the purpose?

Regards,
Simon


More information about the U-Boot mailing list