Hi Linus, 2016-05-31 17:18 GMT+09:00 Linus Walleij <linus.walleij@xxxxxxxxxx>: >> But, this effectively breaks the current DT binding. > > Does it really? What matters is if it still works after the change. Yes. I use syscon_node_to_regmap(). I want to get the regmap of the parent node, not of itself. >> [1] Add the new binding support with a new compatible string. >> (note the old binding is still supported for backward compatibility) >> >> [2] Wait a couple of releases, keeping both of the two bindings >> >> [3] Drop the old binding support >> >> If it is OK, I want to move it forward. > > It is OK with *ME* because I'm just the GPIO maintainer and I'm > not really picky about breaking DT ABI's, I think you need to crack > a few eggs to make an omelet from time to time. > > The problem is rather the devicetree people: do *THEY* accept > this? This case is really an SoC-specific matter. I hope it is acceptable because I am not disturbing other vendors at all. I broke another DT before on my board, https://lkml.org/lkml/2016/2/25/89 in this discussion, Olof said it is a good practice to keep the old binding around, but I do not have to keep it forever. I am not sure how strict devicetree people are, though. -- Best Regards Masahiro Yamada -- To unsubscribe from this list: send the line "unsubscribe linux-gpio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html