Attention is currently required from: Martin L Roth.
Maximilian Brune has posted comments on this change by Martin L Roth. ( https://review.coreboot.org/c/coreboot/+/87185?usp=email )
The change is no longer submittable: All-Comments-Resolved is unsatisfied now.
Change subject: Documentation: Add information about the site-local directory ......................................................................
Patch Set 2:
(1 comment)
File Documentation/getting_started/site-local.md:
https://review.coreboot.org/c/coreboot/+/87185/comment/a5bd75b8_aa5da05e?usp... : PS2, Line 165: One of the most powerful use cases for `site-local` is developing a new : SoC implementation out-of-tree before it's ready to be made public. : This allows you to: : I don't really understand this use case. Even with the points below I don't understand why someone wouldn't just create the new SOC inside the standard SOC directory. When you create a new SOC, you will never have any conflicts with upstream, because the directory doesn't exist upstream.
The other scenario: Assuming the directory already exists upstream and you want to update some things in that SOC and don't want them to be public yet. I still don't understand why these changes would be in the site-local folder, because at some point you will need to rebase onto upstream and rebasing the site-local folder is much harder then just rebasing your patches on top of that SOC.
The only use case I can imagine (and that I have myself) is putting files/patches into the site-local folder for which I know that they are NEVER going to go upstream, because they are just some weird hacks, debug things or binary blobs.
I am not suggesting that this is the only "useful" use case, but before making suggestions to other users I would like to make clear why it is useful to do so and the points below just don't really give me this impression (or at least I don't understand them).