gold work on mingw/mingw64 support

[Ralf Wildenhues]

Hello Vladimir,

* Vladimir Simonov wrote on Thu, Dec 30, 2010 at 06:47:55PM CET:
> The patch's idea is to provide restricted emulation for
> Linux-specific calls, like mremap, mremap, etc. And force
> callers always to keep files in-memory.

> What do you think about this approach and the patch itself?

As a by-stander, I would suggest using configure checks for headers and
functions which might or might not be present, and using #ifdef, if at
all needed, only or at least mostly with those platform-independent
defines set by the configure macros (HAVE_SYS_MMAN_H etc).

One could use gnulib modules to emulate some of the remaining missing
functionality (fcntl FDCLOEXEC, anon mmap).  That would not solve the
remaining mmap issues yet; for those one /could/ write a local gnulib
module, I suppose.

I can see that at least for the current goal each of these steps might
seem like overkill; I'm not sure whether or when they might pay off
sometime later.  Let's see what Ian thinks.

Cheers,
Ralf