Kyösti Mälkki has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/37895 )
Change subject: arch/x86: Top align bootblock program ......................................................................
Patch Set 4:
Patch Set 4:
Patch Set 4:
Aaron, ping. Would you want to introduce new (sub)sections such that id.S and fit.S are properly accounted as sections instead of an arbitrary reserve below .reset?
I think that would be helpful. Ideally we can agree on how to construct that upper end for linking purposes w/o wasting space. Basically make the section small enough (w/ an appropriate alignment) to fit the required entries but not large enough that we don't fragment the space leaving it stranded. We can pursue this path soon-ish, but I don't think we're dependent on it at the moment. Let's wait for Marshall to update his patch stack that includes a normal transit through bootblock and romstage for the new boot architecture. Then we can improve things from there.
The release requirement of !ROMCC_BOOTBLOCK triggered a side-effect when moving from top-aligned to bottom-aligned bootblock, there were some complaints of no longer being able to fit TianoCore payload on some installations with extremely tight SPI flash usage.
This commit targets for removal of a static C_ENV_BOOTBLOCK size. On my loal development trees I have tied the development of reduced-topology static devicetree for the bootblock to this, as well as development for 'CBMEM_CONSOLE-only bootblock, complex-console romstage with replay from CBMEM_CONSOLE'
On your request, this is now put on indefinete hold as well and is waiting for amd/picasso to progress.