I will add my two cents as well. I have already responded in comments
on PR #1651, but I will copy my reponse here form continuity:
I am in 100% agreement with Brennan and Matias. I am very opposed to
this change because it violates all of the architectural principles that
have been established for the NuttX wireless subsystem over the course
of about a 2 year discussion in the old Google group. That architecture
and roadmap are documented here:
https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=139629397
Specifically, the interface to the all complex wireless devices MUST be
through the network layer. Use of character devices in anything other
than the most trivial of radios is strictly forbidden.
Xiaomi cannot contribute this implementation only because they are too
lazy to follow the established architectural principles as required in
the INVIOLABLES.md. This is wrong and cannot be permitted to happen.
I also believe the underlying approach is unethical. It takes some of
the Zephyr code, downloads it at build time, and builds the unauthorized
third party code into the binaries. The purpose of this is to avoid the
NuttX coding standard and to circumvent the ASF's policies for including
third party code. While this approach may meet the "legal" definitions,
it is inherently corrupt at its core and does not meet the "spirit" of
the requirements. This approach is not only lazy, inconsistent with the
NuttX roadmap, but lacks integrity.
This kind of change must NEVER come into the NuttX master branch. This
is not in the spirit of NuttX, it is not consistent with the NuttX
roadmap, custs corners to reduce effort, and lacks integrity. This is
all in contradiction to most of the statements.
