Hi Keith,

On Tue, 12 Sept 2023 at 18:09, Keith Hui <[email protected]> wrote:
>
> Hi Simon,
>
> > Why not CMOS RAM / VVRAM?
> - If we allocate some space in CMOS for console / logging settings,
> then it would allow a similar feature. But it involves changing
> settings on the device. Each board would need to provide some CMOS
> options for this feature as part of the layout file. It would not be
> possible to enable console output without running some code on the
> device to update the CMOS RAM.
>
> (Pardon the formatting; I copied this out of my semi-regular digest)
>

This came in not as a reply to my email, so perhaps others will not see it.

> Many boards already have this space allocated in their cmos.layout.
> Take for example my sandybridge board, asus/p8z77-m:
>
> # coreboot config options: console
> 395          4       e       4        debug_level
> [...]
> # debug_level
> 4     0     Emergency
> 4     1     Alert
> 4     2     Critical
> 4     3     Error
> 4     4     Warning
> 4     5     Notice
> 4     6     Info
> 4     7     Debug
> 4     8     Spew
>
> Setting this to 0 suppresses almost all outputs unless it's a "I'm
> going to die()" scream.
>
> For development/hacking we do want to start with debug outputs on the
> console. If after ample testing a build has proved itself to be
> stable, cmos.default (that is a separate file) in CBFS can be patched
> to change the default log level. This would be done through cbfstool
> (albeit indirectly - extract, patch, re-insert) or a hex editor. I'd
> propose a tool that operates on this binary cmos.layout instead.

But how does that work generically? Doesn't it require boards to have
a particular CMOS layout? Also, how does this work with boards which
don't have CMOS RAM? Finally, is it really correct that booting
coreboot should update the CMOS RAM? What tool is available to adjust
CMOS RAM without having to boot into Linux, etc.? (which may not work)

>
> In short, in my opinion, we don't need a new construct.

I don't know how to join the threads, but I am not convinced, based on
the above.

Another concern is that we don't even enable the serial console at
present in many cases...we need some control of that, separate from
what the debug level is. For example, CBMEM can record all the
console, perhaps at maximum level, but the serial output might want to
be disabled.

Regards,
SImon
_______________________________________________
coreboot mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to