Change in coreboot[master]: libpayload/i8042: Use 'WARNING' instead of 'ERROR' when probing failed

[Nico Huber (Code Review)]

Attention is currently required from: Paul Menzel, Tim Wawrzynczak, Nick Vaccaro, Zhuohao Lee.
Nico Huber has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/60129 )

Change subject: libpayload/i8042: Use 'WARNING' instead of 'ERROR' when probing failed
......................................................................

Patch Set 4:

(2 comments)

Commit Message:

https://review.coreboot.org/c/coreboot/+/60129/comment/9ecc4e57_40824d61 
PS2, Line 13: when doing the autotest.
> Thank you for updating the commit message. […]

It's a minor change so I don't want to block this. But generally this change
is completely wrong. It's wrong to adapt code instead of a test when there
is nothing wrong per-se with the code. Especially when the test tool can't
be told what to expect. Also, that this is designing upstream code by a
downstream (IMHO broken) tool raises further questions: What if somebody
else has a downstream tool that checks for ERROR in case they test the
absence of the controller?

This change is wrong and the reasoning rather arrogant.

Commit Message:

https://review.coreboot.org/c/coreboot/+/60129/comment/59ae1615_cfb7d220 
PS4, Line 9: should
IMO we shouldn't say "should" unless we can explain why. That
some test doesn't count as explanation because tests are not
supposed to tell us what to do.