Attention is currently required from: Angel Pons. Michael Niewöhner has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/48383 )
Change subject: mb/fujitsu/d3410-b1: Add new mainboard ......................................................................
Patch Set 12:
(4 comments)
File src/mainboard/fujitsu/d3410-b1/bootblock.c:
https://review.coreboot.org/c/coreboot/+/48383/comment/08192070_919b0bc9 PS12, Line 9: We had a private discussion on this already, so just for the record:
1. IMO all pads should be configured, even if they match the "reset" values, because the GPIO file also serves as sort-of documentation how the pads are configured.
2. Reset values: There is a possibility to change these "reset" values using the FIT tool. On ME-platforms only a few pads can be configured, while on server platforms (SPS) many more pads' reset values/functions can be edited via dropdowns.
That means, the values in the datasheet is the hardware reset value but software (probably ME) can override them and we can't always rely on these values and should instead always configure all pads, so we have a defined, sane state.
https://review.coreboot.org/c/coreboot/+/48383/comment/23e3539a_41f64d1f PS12, Line 11: UP_20K uhm, that is 0b1100, which is only valid in eSPI mode. LPC *must* have `NATIVE` (0b1111) here, according to the pch ds vol2. does that board use eSPI?
https://review.coreboot.org/c/coreboot/+/48383/comment/7875d858_6f296347 PS12, Line 34: DEEP inteltooldump says 0b11 (reserved), just like the GPP_I* below
https://review.coreboot.org/c/coreboot/+/48383/comment/932be254_7983b97c PS12, Line 57: PAD_CFG_NF(GPP_I0, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I1, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I2, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I3, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I5, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I6, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I7, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I8, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I9, NONE, PWROK, NF1), : PAD_CFG_NF(GPP_I10, NONE, PWROK, NF1), : your inteltool dump says PADRSTCFG=0b11, which is "Reserved"... eds also says reserved... hmm. default for these pads is DEEP, which should be ok