Nico Huber has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/35884 )
Change subject: soc/intel/Kconfig: Move MMCONF_BASE_ADDRESS out of a common place ......................................................................
Patch Set 1:
(2 comments)
https://review.coreboot.org/c/coreboot/+/35884/1//COMMIT_MSG Commit Message:
https://review.coreboot.org/c/coreboot/+/35884/1//COMMIT_MSG@10 PS1, Line 10: This requires soc/intel target to : explicitly set this in the SOC dir
It's a bit like the overuse of __weak in soc/intel to make stuff build without being correct. […]
To be clear, I'm just bikeshedding the relation to the previous sentence: if the fact that you added the compile time error would really require this change, Jenkins wouldn't have agreed on the original change :-p
What I read between the lines is roughly that "[To be really useful] this requires [...]".
https://review.coreboot.org/c/coreboot/+/35884/1//COMMIT_MSG@12 PS1, Line 12: wrong
How can it be wrong? does the FSP binary depend on a specific […]
I agree but still wonder: With a change like this, that has no technical requirement, how can we prevent that it's not turned back. e.g. somebody sees there are 8 out of 9 platforms that use the same value and puts it back into common/...
I don't like comments, but in this case it could help to add something like
Needs to be what individual FSP versions expect. Keep this in the individual soc/ directory.