Nico Huber has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/34287 )
Change subject: soc/intel/{cnl,icl}: Always use CAR NEM enhanced by default ......................................................................
Patch Set 2: -Code-Review
(1 comment)
I've just boot-tested all affected boards (empty set, assuming abuilds configurations). So I don't think we should put more effort into this.
However, on the FSP-T integration side, I've learned by now that my initial hunch why my board isn't booting must have been wrong: I thought it was because I left CPU_MICROCODE_CBFS_LEN/_LOC at 0. But the FSP integration guide for CFL (what I tested) says this would be ok.
So it would be really nice if somebody could confirm if the FSP-T integration works/ever worked upstream.
https://review.coreboot.org/c/coreboot/+/34287/2/src/soc/intel/cannonlake/Kc... File src/soc/intel/cannonlake/Kconfig:
https://review.coreboot.org/c/coreboot/+/34287/2/src/soc/intel/cannonlake/Kc... PS2, Line 272: default USE_CANNONLAKE_CAR_NEM_ENHANCED
The default in a choice is to pick the first entry, so in this case, there's a default, just no stat […]
I just made some rough statistics of current occurences:
$ # overall choices: $ git grep "^choice$" | wc -l 125
$ # choices with conditional default: $ git grep -A2 "^choice$" | grep $'\t''default [^[:space:]]+[[:space:]]' | wc -l 13
$ # choices with unconditional, explicit default: $ git grep -A2 "^choice$" | grep $'\t''default [^[:space:]]+$' | wc -l 58
So I'm surprised to see that stating the obvious is already the prevailing pattern: 58:54 (57:54 excluding this patch :-P). If somebody wants to bikeshed my statistics: just remove the `| wc -l` at the end and you'll see that there are some clusters, e.g. if you'd remove payloads from the statistics the result would look much different ;)