On 08/20/2012 11:54 AM, Mark Brown wrote: > On Mon, Aug 20, 2012 at 11:49:11AM -0600, Stephen Warren wrote: > >> The problem there is that will prevent any regulators from being >> registered at all, until the patch is merged with the regulator >> tree. That will also prevent PCIe working on Harmony. Those both >> work right now, so that's a bisection breakage. > >> I suppose the solution here is to merge the regulator tree into >> the Tegra branch before the Harmony regulator patch. Mark, can we >> do that? > > I've no idea what you're talking about here, sorry. What are > "this" and "the patch"? Surely you can just add whatever DT > changes you need to add without the thing that uses the bindings? When booting Harmony using device tree, there's a special case that registers all the regulators using the remnants of the Harmony board file. Hence, regulators work right now even without the DT instantiating them. The patch in this thread adds all the required regulators to the DT file. However, it should also remove the special case so the regulators don't get registered twice, once from the board file and once from DT. However, doing that would prevent regulators getting registered at all, since v3.6-rc* don't have all the required (TPS6586x) regulator (driver) patches to get the regulators instantiated from DT. Everything will only work in linux-next or sometime during the 3.7 merge window. Hence, git bisect would be broken. One solution to this is to bring the regulator tree into the Tegra tree as a dependency, and then apply a patch which adds the regulators to DT, and removes the special case from the board file. I'm asking for your OK to do that. -- To unsubscribe from this list: send the line "unsubscribe linux-tegra" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
