On Saturday 03 March 2012, Thierry Reding wrote:
> * Thierry Reding wrote:
> > * Arnd Bergmann wrote:
> > > On Thursday 23 February 2012, Thierry Reding wrote:
> > > > > Is this only used for the device tree functions? If so, I would
> > > > > recommend
> > > > > making it less generic and always search for a device node.
> > > >
> > > > It is currently only used to look up a struct pwm_chip for a given
> > > > struct
> > > > device_node, yes. But I can see other uses for this. For instance this
> > > > could
> > > > be useful if we ever want to provide an alternative way of requesting a
> > > > PWM
> > > > on a per-chip basis.
> > >
> > > Nah, just keep it simple for now. If we need it later, we can still
> > > add something like this back, but for now it's just a source of
> > > confusion and possible bugs.
> >
> > Will do.
>
> I turns out that this is not as easy to do as I thought. The problem is that
> if I remove the pwmchip_find() from the core and move the lookup
> functionality into the OF support code I no longer have access to the list of
> PWM chips. So I guess it will have to stay in to keep things encapsulated
> properly.
Well, or you move everything into the one file. If you have the device tree
support function in the same file as everything else, you no longer need
the additional complexity.
Arnd
_______________________________________________
devicetree-discuss mailing list
[email protected]
https://lists.ozlabs.org/listinfo/devicetree-discuss
