Hung-Te Lin has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/46442 )
Change subject: mb/google/kukui: fix SKUID config for burnet/esche ......................................................................
Patch Set 1:
(1 comment)
https://review.coreboot.org/c/coreboot/+/46442/1/src/mainboard/google/kukui/... File src/mainboard/google/kukui/boardid.c:
https://review.coreboot.org/c/coreboot/+/46442/1/src/mainboard/google/kukui/... PS1, Line 99: if (CONFIG(BOARD_GOOGLE_BURNET)) { : cached_sku_id = (0x10 | get_adc_index(SKU_ID_CHANNEL)); : return cached_sku_id; : } :
The LCM_ID HW strap of kappa & juniper are actually 1. Their sku-ids are 0x10 (not 0x0) now. I'm worried about changing the rule of sku-id for MP devices.
Yes but they don't have other SKUs (single SKU) so it's probably not a big deal - especially there is only one DTS for them.
In the meantime we can also hard-code LCMID to 1 for JACUZZI devices.
I'd recommend that we still focus on debuging the floating HW strap value of LCM_ID. Then, we don't have to patch anything.
That's also true, but I was concerned maybe juniper and kappa have exactly the same issue (floating LCM_ID) and we didn't notice it (due to single SKU, no matching issue). In that sense hard-coding the LCM may also be a good idea for future.
Conclusion: I think we should keep debugging, and maybe also check shipped devices as well. And if that is a hardware issue that cannot be easily fixed, we should consider putting a hard-coded value to JACUZZI LCM_ID.