* Rich Felker <dalias@libc.org> [2016-05-05 22:44:54 -0400]:
> not built at all for a c/c++ only build. Other langs are completely
> untested and probably don't work, but it would be nice to find out why
> and fix the ones that are practical to fix.

btw, for fortran --disable-libquadmath is not enough,
you need --disable-libquadmath-support too, see

https://gcc.gnu.org/install/configure.html

so i think those options should be recommended
together.

there are patches to make java and ada frontends
work (by alpine linux) but those are much less
tested than the c,c++ case.

> > Are you planning on supporting GCC 5.3? I may be able to contribute
> > some patches.
> 
> I suspect the same patches for 5.2 apply just fine, but I haven't
> taken the time to test yet. If you can confirm they work or tweak them
> so they work I'd be happy to add 5.3 support. Other versions I'd like

i think the powerpc patches wont apply, but
i have all my patches rebased for 5.3

> to support are 4.7.4 (last C-only, also last before some sketchy

likewise.

> optimizations were added) and 6.1.0, but in order to add a version it

likewise.

> should have proper support (at least the main musl patch with dynamic
> linker names, include paths, etc. fully ported) rather than just being
> enough to compile.

but i didnt test these patches with mcm, they don't have the
pie/fdpic/sh stuff and they might contain extra changes that
probably not everyone wants.