because "wrong" casts will either be optimized away or never performed.

libraries. Old .sl extension works just fine, but it .so which is default.

Win64 comes in two flavors, IA-64/Itanium and AMD64/Opteron. The
suggestion is to refer to former as WIN64I and latter - WIN64A

Apparently, the length *including* the NUL byte should be used.

Contributed by Andy Polyakov <appro@fy.chalmers.se>

Make a nicer comment, as we don't really know for sure that it's
really needed, and just want to play on the safe side.

Suggest by Andy Polyakov <appro@fy.chalmers.se>

with HP-UX offering 14 for NAME_MAX?

SIXTY_FOUR_BIT_LONG) were failing to pass 'cd test; make test_bn'.

Some code beautification.

Change the macro CP_THREAD_ACP to CP_ACP, because the latter is more
widely defined.

Add a conditional macro definition in case FindFirstFile and
FindNextFile aren't properly defined (might happen on WinCE).

Suggested by Andy Polyakov <appro@fy.chalmers.se>

Windows changes that detects if multibyte characters are available and
deals with them properly.

Contributed by Andy Polyakov <appro@fy.chalmers.se>

NUL-teminated at all times, and that we don't make unneeded calls to
free().

instead of local time.

I also used this opportunity to clean up some out-of-date targets and
re-group targets by OS.

we were mislead by _MSC_VER macro, which is defined by *all* Windows
Microsoft compilers.

Submitted by: Nils Larsch

(suggested by Adam Young <ayoung@cigital.com>)

Submitted by: Nils Larsch

(incidently, this also tells VMS that there exists a new symbol in the
SSL library)

we need to have them among the symbols that should appear in the
transfer table, at least on VMS (and it wouldn't surprise me if
Windows would whine as well).

Lucky me, I had prepared for this :-).

identities disabled.

Don't have a constant that you're going to assign to, that's just
plain stupid (I was the stupidhead here...).

linux-x86_64 target.