* Russell King - ARM Linux wrote: > On Tue, Feb 07, 2012 at 08:04:00AM +0100, Thierry Reding wrote: > > One of the major problems when converting to a non-global namespace is how > > to > > represent the relationship in code. For device tree this should be easy to > > do > > because it has all the infrastructure in place. For non-device-tree devices > > I > > have no idea yet how this could be done. Perhaps by using something like the > > clock API and using names for lookup? > > Well, the first step is to make sure that the pwm 'chip' drivers don't > know anything about the global numberspace. > > That limits the extent of the global numberspace to the pwm core and > pwm users. You then don't have to worry about the pwm chip drivers > while you work out how to solve the other half of the problem.
I see. That would mean that pwm_request() gets a global PWM index, looks up the corresponding PWM chip and device then pass the chip-relative index to the driver's ops. I think that can be simplified even further. For instance if the pwm_device structure contains a pointer to its parent struct pwm_chip, then that can be used directly to call into the driver's ops. And the ops could be modified to take a pwm_device instead of the global PWM index. That would be similar to how struct irq_data is used. Thierry
pgpRJqVeNdq9E.pgp
Description: PGP signature
_______________________________________________ devicetree-discuss mailing list [email protected] https://lists.ozlabs.org/listinfo/devicetree-discuss
