AFAIK there has never been any IBM marketing request to the libvirt project
in respect of CPU names. Perhaps that was communicated to IBM contributors
writing patches in private, but from POV of the libvirt community upstream
that is irrelevant. We review patches based on their technical characteristics
and AFAIR at the time QEMU became case insensitive, libvirt decided to stick
with uppercase names, driven by the historical precedent that we had already
established with earlier versions of libvirt.


In power10 it seems like IBM marketing changed its mind [1] and now the model
name is camel case, so now we're supposed to call them Power10, Power11, etc.
This is the prime example of why marketing should be ignored when making
technical decisions. They will always change their minds every few years.

Ignoring case seems a sensible change to make since we do not want
to be hostage of IBM marketing whims.
I'm not sure we have any significant precedent of ignoring the case of user
specified input. The general goal of libvirt XML input is to specify the
data in a single canonical format and avoid redundant ways to specifying
the same thing.

IMHO we have two options - continue to use uppercase in all ppc CPU models,
or switch to lowercase only for new CPU models we add.

With regards,
Daniel