Hi Pavel, Am 27.04.2015 um 22:35 schrieb Pavel Machek <pa...@ucw.cz>:
> Hi! > >>> In my opinion making something a node in / is always the >>> last resort and in my perspective it has been handled in such a way >>> so far. >> >> But that contradicts some documents I have found and linked. Please >> show me a document about DT that supports your view. >> >> I agree that both views can be valid, but in lack of some ?official? >> guideline >> we can?t decide ourselves. > > This is not how kernel development works. > > We can decide ourselves, and DT people will swear at us if we decide wrongly, > and then Linus swears at them if they do too obvious mistakes. > > Lets KISS. some time ago you also wrote in this thread: > Device tree is not operating system specific. In other words: this is a discussion about compatibility with the world outside of Linux. And compatibility can not be achieved by decisions about Linux implementation details, even if Linus is deciding. He can just decide to stay incompatible (but then it should IMHO be a general decision and not case specific). So I think we should follow some basic design principles, if they exist (which I don’t know), to have this compatibility. And then adapt the Linux implementation (using the KISS principle). Anyways I have not yet seen any comment from the DT people or did I miss them? BR, Nikolaus -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
