On Wed, Jul 21, 2004 at 01:00:51PM +0200, Jesper Louis Andersen wrote:
> Some compilers treat the inline keyword as a force. They will, how
> bad it might be, inline a function named as such. Other compilers
> totally ignore the inline keyword. And others again will let the
> inline keyword affect their decision heuristic.

There's some general misunderstanding about what inline means.
If you look at the discussions on the gcc mailing-lists about inline,
and specifically static inline/extern inline and stuff like that, you'll
see that even a guy named Linus Torvalds got it completely wrong...

inline has always been a hint to the compiler, and nothing more.

The only annoying issue with gcc and inline semantics is that some
asm constructs require inline to work right...