Marshall Dawson has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/37485 )
Change subject: arch|cpu/x86: Add Kconfig option for x86 reset vector ......................................................................
Patch Set 3:
(3 comments)
I am late for the review... -1 from me.
I've been meaning to ask what you'd like to do with this change.
https://review.coreboot.org/c/coreboot/+/37485/3/src/arch/x86/failover.ld File src/arch/x86/failover.ld:
https://review.coreboot.org/c/coreboot/+/37485/3/src/arch/x86/failover.ld@17 PS3, Line 17: rom : ORIGIN = CONFIG_X86_RESET_VECTOR - 0xfff0, LENGTH = 64K
This file is for romcc bootblocks only, I think you should have just leave this untouched. […]
I'd missed the one below. I'm OK with making the one above hardcoded again.
https://review.coreboot.org/c/coreboot/+/37485/3/src/arch/x86/id.ld File src/arch/x86/id.ld:
https://review.coreboot.org/c/coreboot/+/37485/3/src/arch/x86/id.ld@16 PS3, Line 16: .id (.): {
I thought ID_SECTION_OFFSET is "defined" to be offset from the end of non-volatile memory and not re […]
You might be right. I looked back to ~2012 and didn't see much in the way of a definition.
Having this here currently ensures the linking works properly when picasso's build is bootblock/romstage. However, I could #if this file out memlayout.ld when RESET_VECTOR_IN_RAM.
https://review.coreboot.org/c/coreboot/+/37485/3/src/cpu/intel/fit/fit.ld File src/cpu/intel/fit/fit.ld:
https://review.coreboot.org/c/coreboot/+/37485/3/src/cpu/intel/fit/fit.ld@16 PS3, Line 16: .fit_pointer (.): {
This should never move?
I believe that's right, and X86_RESET_VECTOR should always be fffffff0 in Intel systems. This file is only included when CONFIG(CPU_INTEL_FIRMWARE_INTERFACE_TABLE). I'd be fine hardcoding it again.