On Sun, Jul 19, 2020 at 8:36 PM Andy Shevchenko <andy.shevchenko@xxxxxxxxx> wrote: > On Thu, Jul 16, 2020 at 6:17 PM Rodolfo Giometti <giometti@xxxxxxxxxxxx> wrote: > > On 16/07/2020 15:38, Linus Walleij wrote: > > > I see but this interface is not designed for such complex usage nor to compete > > with the current character interface! It is designed to allow boards > > manufactures to "describe" some I/O lines that are not used by any driver in the > > device tree, > > Why are they not in firmware tables? Platform is a set of hardware > that makes it so. > If something is not in DT, then there is no possible way to know what > is that line? > > Or in other words how does the OS know that the certain line is > connected to a relay? IIUC Rodolfo's idea is to provide this with a DT compatible. The use case will be industrial automation-ish tasks from userspace. Currently the only mechanism we have in the device tree to assign a use for a line is the "gpio-line-names" property, which creates a name that is reported upward to the character device. Rodolfo's patch is for scripting use cases, assigning some lines for some cases to be handled by scripts, not the character device. What I am a bit worried about is if this would be a Linuxism, as DT should be OS neutral. Yours, Linus Walleij
