On Thu, Nov 01, 2012 at 03:01:28PM +0100, Linus Walleij wrote: > On Thu, Nov 1, 2012 at 1:07 PM, Mark Brown > > > For the pin hogging I'd actually been thinking separately that we should > > just have the device core do a devm_pinctrl_get_set_default() prior to > > probing the device and store the result in the struct device. That > What if I retrieve this in the pinctrl subsystem using > bus notifiers and then expose the struct pinctrl * to > the clients by using pinctrl_get() and when you get > such a handle in your probe() you know that the > pinctrl subsystem has already fetched the handle and > set it to "default" at that point? I'm not sure I parse a problem from the above? > I just worry whether there is a fringe case where the default > state is not be be default-selected in probe(), the API > semantics does not mandate that. But I think this is the case > for all in-kernel drivers so we wouldn't be breaking anything > by doing this right now. And platforms can just leave the > "default" state undefined in that case. Yes, that had been my thinking too though I'd really expect that the platform ought to be able to think of something sensible to do by default. > Then what to do with sleep and idle is a question we need > to handle next. Requiring PM domains for this is one > approach, albeit a bit controversial. Yup. Notifiers are another option again I guess. There's far fewer drivers doing anything at all with that so it's a bit less urgent.
Attachment:
signature.asc
Description: Digital signature
