Attention is currently required from: Benjamin Doron, Felix Held, Lean Sheng Tan, Matt DeVillier, Maximilian Brune, Philipp Hug, ron minnich.
Julius Werner has posted comments on this change by Benjamin Doron. ( https://review.coreboot.org/c/coreboot/+/84796?usp=email )
Change subject: lib/{fit,fit_payload}.c: Enhance support for FIT images ......................................................................
Patch Set 10:
(3 comments)
File payloads/Kconfig:
https://review.coreboot.org/c/coreboot/+/84796/comment/89b7803b_cfe6c752?usp... : PS9, Line 186: default y if PAYLOAD_LINUX && (ARCH_ARM64 || ARCH_RISCV || ARCH_ARM)
Oh, I think that made it align with other instances of this dependency in the file, which allowed me […]
Acknowledged
File src/lib/fit_payload.c:
https://review.coreboot.org/c/coreboot/+/84796/comment/bb528311_24a2157d?usp... : PS9, Line 29: printk(BIOS_ERR, "WEAK: %s:%s() called to add %s to FDT\n", __FILE__, __func__, name);
Do you have any ideas for paths? Possibly, `/firmware/secondary-images`. If we decide on some path and a PR for the spec is up, I'll change this code to reflect that and drop the callout.
You mean in the FDT? That's the problem, the FDT is really specific to the payload you're loading, there are no "universal" rules on paths. `/firmware/coreboot` and all that stuff is specific to the Linux kernel. UPL has its own spec on how the FDT is supposed to look like, which tries to vaguely match the kernel in style where that works but is really not the same. If you say that there are other firmware payloads that are loaded by this but are not UPL, then I don't know how the FDT is supposed to look like for them, that's up to them. The entire point of UPL was to standardize that FDT, after all.
No, see section 4.1 of the FIT spec.
You mean the one that only goes up to section 3.4 here: https://fitspec.osfw.foundation/ ? ;)
It sounds like I need some extra context that's not yet officially published(?) and maybe then I'll have an easier time following what all this is supposed to do.
https://review.coreboot.org/c/coreboot/+/84796/comment/0611fae8_89a52d48?usp... : PS9, Line 403: &fdt, &initrd
Yes. […]
Oh, okay, I see how the FDT thing gets added in the later patches. But for initrd, if one specific UPL payload had an initrd, wouldn't that just be just another secondary loadable for it? Wouldn't `load_secondaries()` take care of it?
-
Julius Werner (Code Review)