Change in coreboot[master]: cpu/x86/smm: Add log_level that can be overridden for SMM log level

[Nico Huber (Code Review)]

Attention is currently required from: Marc Jones, Patrick Rudolph, Jonathan Zhang, Johnny Lin, Rocky Phagura, Jingle Hsu, Angel Pons, Arthur Heymans, Morgan Jang, Kyösti Mälkki.
Nico Huber has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/49460 )

Change subject: cpu/x86/smm: Add log_level that can be overridden for SMM log level
......................................................................

Patch Set 9:

(2 comments)

Commit Message:

https://review.coreboot.org/c/coreboot/+/49460/comment/7e87d9ec_4697336f 
PS9, Line 16: for SMM.
I couldn't find any discussion about this issue. Sorry if I just
missed it somewhere else.

Please provide more details. For instance, is it too much time
for every SMI? or is it generally too much time, even if you'd
only query VPD once during boot in SMM?

Patchset:

PS9: 
I don't understand why OVERRIDE_CONSOLE_LOGLEVEL is
used to guard here. Your VPD/SMM situation is not the
only reason why it might be set. So guarding or not, you'd
add the code for builds that wouldn't make use of it. That
you need the `weak` declaration makes that obvious.

There's a lot of #if in .c files, please avoid that.

But maybe first, we should talk about your requirements;
see inline comment. Probably a dumb question: Can't you just
cache the loglevel in your get_console_loglevel() after its
first invocation? Then the VPD code would only run once. Is
that bad already?

If you really need to pass the loglevel from ramstage, I
think you should do either of these two:

o Make passing the loglevel to SMM an option for everyone.
Without the need for additional code.

o Use a generic mechanism to pass board-specific information
into SMM. That it's about the console shouldn't be visible to
the SMM-loader code.

But don't mix these things.