Hi Tom, On Tue, 5 Oct 2021 at 09:52, Tom Rini <tr...@konsulko.com> wrote: > > On Tue, Oct 05, 2021 at 09:33:18AM -0600, Simon Glass wrote: > > Hi Wolfgang, > > > > On Tue, 5 Oct 2021 at 08:56, Wolfgang Denk <w...@denx.de> wrote: > > > > > > Dear Simon, > > > > > > In message > > > <CAPnjgZ1tRi5SsqU0K8HXgj-4xCs7i9TLX4Mj0_D=cpj8ban...@mail.gmail.com> you > > > wrote: > > > > > > > > > > Add a feature that brings in a .env file associated with the board > > > > > > config, if present. To use it, create a file in a board/<vendor>/env > > > > > > directory called <board>.env (or common.env if you want the same > > > > > > environment for all boards). > > > > > > > > > > This should be no exclusive "or" here. If a common.env exists, it > > > > > should be used for all boards, and if additionally one ore more > > > > > <board>.env files exist, these should ALSO be applied to the > > > > > respective boards. > > > > > > > > Is it not enough to use #include in the main file? We have a similar > > > > feature with the u-boot.dtsi files and in that case we only choose the > > > > most specific. > > > > > > 1) This requires that the .env files are run through CPP, which is > > > only added in a later patch. > > > > OK perhaps I should just merge the patches. It is a bit artificial > > having two and it seems that people agree we need the += syntax. > > It's important to maintain bisectability, yes. But functionality > should be evaluated at the end of the series, not intermediate steps. I > don't have a strong opinion either way on if these two patches are > merged, or not. So on a similar note, all of the feedback about the > current env documentation is good and helpful, but I think a txt -> rST > then enhance the rST makes the most sense so that we don't "hide" > improvements within the migration.
OK. I added Wolfgang's comments in the patch where he made them but will split them out. Regards, Simon