Attention is currently required from: Lance Zhao, Tim Wawrzynczak, Angel Pons. Kyösti Mälkki has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/51638 )
Change subject: ChromeOS: Separate NVS from global GNVS ......................................................................
Patch Set 4:
(4 comments)
File src/acpi/gnvs.c:
https://review.coreboot.org/c/coreboot/+/51638/comment/a7c6c8ae_c841e5d7 PS4, Line 43: chromeos_init_chromeos_acpi
am I missing something? doesn't this patch effectively drop generating chromeos nvs?
Looks like I somehow removed [WIP] with rebase.
The question is if we should split the initialisation or not.
File src/vendorcode/google/chromeos/gnvs.c:
https://review.coreboot.org/c/coreboot/+/51638/comment/0f2b0370_438d7d7d PS4, Line 33: static __unused void chromeos_init_chromeos_acpi(void)
Who should call this?
That's the question, should the call be inside acpi/gnvs.c which creates dependency on ACPI_SOC_NVS which is inaccurate. (Does not really hurt though).
https://review.coreboot.org/c/coreboot/+/51638/comment/929657d2_d05bfd77 PS4, Line 38: chromeos_acpi = cbmem_add(CBMEM_ID_ACPI_GNVS, sizeof(struct chromeos_acpi)); Needs new ID.
https://review.coreboot.org/c/coreboot/+/51638/comment/e5460264_8ae02315 PS4, Line 96: static __unused void acpi_fill_cnvs(void)
Who should call this?
I was thinking inside vc/google/chromeos/acpi.c:chromeos_dsdt_generator(). Then I realised that calling chromeos_dsdt_generator() from inside mb/ is somewhat pointless as the implementation is not tied to passed dev pointer. And should we query the acpi name with that node, we would not get to Device (CHRW) that this NVS is related to.