Attention is currently required from: Arthur Heymans, Felix Held, Julius Werner, Jérémy Compostella, Karthik Ramasubramanian, Martin Roth, Paul Menzel, Simon Glass.
Simon Glass has posted comments on this change by Simon Glass. ( https://review.coreboot.org/c/coreboot/+/77712?usp=email )
Change subject: Introduce a coreboot Control Block (CCB) ......................................................................
Patch Set 13:
(5 comments)
Patchset:
PS10:
My concern is about *not* adding the option that causes so many issues and requires this to be so co […]
Just accessing FMAP or CBFS produces console output, so it is not just one stage. Each stage would produce output before the banner if we don't use the cache and CBMEM. This series resolves that in a nice, clean way.
Stacking early CBMEM output is itself an added complexity. Why not just solve the problem properly?
As you can see, putting CCB in bootblock is pretty tidy.
PS10:
You keep saying that but you're not really explaining why that is true. […]
Would you mind downloading the patches and trying them out? It might help you understand the trade-offs better.
ChromeOS had this feature 10 years ago on Snow, but it was lost somehow along the way.
A silent console needs to be silent. It can't output a small number of lines just because it is annoying to suppress those. Nor can it suppress the bootblock banner in 'loud' mode. We just need to know, right at the start, whether the console is silenced or not.
File src/arch/x86/car.ld:
https://review.coreboot.org/c/coreboot/+/77712/comment/b070a783_cba4de04?usp... : PS10, Line 45: CCB(., 0x10)
Yes but timestamps have been there for decades and are widely used by the majority of coreboot users […]
Open source projects can change and evolve. I actually feel that coreboot has lagged a bit in this area. It could really do with some updates, and a more positive attitude to genuinely new features, IMO.
File src/lib/ccb.c:
https://review.coreboot.org/c/coreboot/+/77712/comment/8d0ea4f0_f742e4c3?usp... : PS12, Line 82: BUG()
return NULL after this to avoid returning an uninitialized ccb variable?
Done
File src/lib/hardwaremain.c:
https://review.coreboot.org/c/coreboot/+/77712/comment/f8ab1d33_bd40e694?usp... : PS10, Line 449: cbmem_initialize();
No, but the console still won't start printing until `console_init()` is called, even if it is enabl […]
Well you can disable CCB in that case. Once it is debugged you can re-enable it. What am I missing?