>Except, that i don't think that the strategy of:
>	"make (at least parts of) it build with the native compiler,
>	then switch to gcc when we've shown that something useful can
>	happen"
>is at all uncommon when doing ports...

Yes, but how much would we lose by just starting with gcc?  On a "modern"
machine it appears to take 6-8 hours, maximum, to build gcc, including
c++, objective C, and Fortran.  Now, older machines may come up, but even
then, it's possible to build.

>Now, granted, DEC's compiler is "ANSI-enough" (it supported
>prototypes, but unless i'm mistaken, at that time didn't define
>__STDC__, or relied on OSF/1 user-land headers to do so), but that may
>not be true for other ports.

Are we anticipating any new systems that are not supported by gcc?  Obviously,
it'd be nice if NetBSD compiled on everything with whatever native compiler.
It'd also be nice to switch to new-style function declarations, which are
distinguished from old in that they are *not* an obsolescent feature.

I doubt this would make things much faster, but it would enable more
potentially useful warnings, and it would make us less dependant on gcc
in particular, and more dependant on ANSI.  I consider this a potential
win.

I know this comes up occasionally, but can anyone point at a likely or
desireable port that will not have an ANSI compiler?  It's looking like
ANSI compliance is getting to be more and more the default state of
compilers.  (New ones.  But we've already got gcc running under every
old one that seems relevant.)

-s