@duncan-roe Sorry for reacting late, but I'm not really sure which problem you are addressing. Is this for C code? An inline function in C++ does not need to be static. I don't think this is a frequently asked question (I don't remember it being asked before).

Yes for C. An inline function declared in a header has to be static else ld complains (the 2nd time the header is included IIRC). static inline is absolutely the normal way to declare public inline functions - see for example the Linux source tree.
I do think it's worth documenting how to deal with static inline functions.
In #11339 I had this documentation as a config example, but @albert-github suggested to make it an FAQ.

For context (I'm posting this just because, as a C++ developer, I was surprised about this C behavior):

• "C99" section in "Inline (C and C++)" Wikipedia article
• "GNU inline vs. C99 inline" by Josef “Jeff” Sipek
• "Different semantics for inline functions" subsection in "Porting to GCC 5"
• "inline function specifier (C documentation)" on cppreference.com
• Answer on Reddit on why adding just inline may be not enough