On Wed Jul 15, 2026 at 5:06 PM CEST, Bartosz Golaszewski wrote: > On Wed, 15 Jul 2026 15:52:07 +0200, Bartosz Golaszewski <[email protected]> > said: >> On Mon, 13 Jul 2026 20:45:13 +0200, Danilo Krummrich <[email protected]> said: >>> On Mon Jul 13, 2026 at 5:11 PM CEST, Bartosz Golaszewski wrote: >>> >>> So, either we have to special case platform_device_set_of_node() too (for >>> the >>> case that a swnode is set already), or just require that the function must >>> only >>> be called when no node has been set, as all users already do; I'd go for the >>> latter. >> >> This sounds good too and it should be easy to implement another test case >> while >> at it. > > Oh but wouldn't that require us to introduce an integer return value for these > functions? Or are you talking about simply updating the API contract?
Yes, we could update the API contract; we could also consider adding a WARN_ON() for the case a node is already set. > I think calling device_remove_software_node() on > is_software_node(dev_fwnode(dev)) in platform_device_set_fwnode() be enough > to handle that case, right? Yes, we can special case it, but the "replace the existing node if any" part of that function is already dead code. So, if no one needs it now, I'd rather remove it and wait for someone to have a valid use-case.
