On Mon, Jan 30, 2017 at 9:14 PM, David Gibson <david@xxxxxxxxxxxxxxxxxxxxx> wrote: > On Tue, Jan 24, 2017 at 11:45:31AM -0600, Rob Herring wrote: >> While '#', '?', '.', '+', '*', and '_' are considered valid characters, >> their use is discouraged in recommended practices. >> >> Testing this found a few cases of '.'. The majority of the warnings were >> all from underscores. > > Hmm. The Opal firmware on POWER8 machines uses both '.' and '#' in > node names in some places. So I'm not terribly convinced this is a > good idea. When would it be using a flat tree and dtc? I know the DT is more dynamic on these systems, but it would be nice if a dts was published as reference. These checks are more something I want to discourage new uses of, not necessarily fix existing cases. I've thought about making the warnings be levels rather than true/false. Or perhaps something like -Wall. Then these checks can be off by default. The kernel build is doing that already for the unit address checks. Rob -- To unsubscribe from this list: send the line "unsubscribe devicetree-compiler" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
