Attention is currently required from: Martin L Roth.
Keith Hui has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/76963?usp=email )
Change subject: mb/samsung/lumpy: Clean up MRC PEI data ......................................................................
Patch Set 1:
(2 comments)
File src/mainboard/samsung/lumpy/early_init.c:
https://review.coreboot.org/c/coreboot/+/76963/comment/1ce9b32a_2210fac1 : PS1, Line 118: uint8_t spdaddr[] = {0xa0, 0x00, 0x00, 0x00}; : uint8_t tsaddr[] = { 0x30, 0x00, 0x00, 0x00}; : uint16_t usbcfg[16][3] = {
I won't comment on the rest of the patches, but I think these data structures should be marked as co […]
Will do for next patch set.
https://review.coreboot.org/c/coreboot/+/76963/comment/365d1a8c_02dfd7ad : PS1, Line 137: pei_data->system_type = 0; // 0 Mobile, 1 Desktop/Server
How did you determine what is needed for each platform? On the last patch for the lenovo x220, this […]
I can only verify for p8z77-m as I own no other hardware. system_type in particular is determined in raminit_mrc.c by some other function. This is where I hope the community comes in and help boot test them. I can only think of adding a debug printk() before and after the change and see for ourselves.
Some other settings are easier to determine:
ec_present is simply copied to devicetree, omitted if 0. max_ddr3_freq is double of max_mem_clock_mhz in devicetree. This will move to devicetree. If gbe_enable is 1 but gbe device is not enabled in devicetree, the line stays and someone else should reconcile things. pcie_init is whether peg10 device is enabled in devicetree; northbridge code also knows to not enable it if it detects a sandybridge CPU. Boards up to this point just hard code it.
-
Keith Hui (Code Review)