[Mingw-users] Newbe-ish questions on Mingw vs cygwin

[Paul K G. <pk...@gl...>]

I've been using cygwin/gcc to port some software originally written using
VC++.  I was having some issues with OLE and found a few things to help.
Over the course of this the cygwin mailing list said I should pick this up
on the mingw list.  Also, one of the solutions for me is to compile with
the -fvtable-thunks flags.  This mean new copies of libgcc and libstdc++,
which I now see are uploaded (via an announcement on this list).

This gets to my questions:

1)  What is the relationship between cygwin and mingw?  I've seen that
using -mno-cygwin effectively gets me mingw, which is fine with me, but if
this is the case then where should I install the re-compiled gcc libs?  I
have not been using mingw itself for my compiles, but cygwin with the
no-cygwin flag.  How do these two system interweave, if at all?  The faq
only seems to state that mingw is an offshoot of cygwin, but doesn't cover
interoperability.

2)  I've been trying to understand the -fvtable-thunks flag.  The quoted
section of gcc.info does not bring understanding to me.  Is it simply that
there are several ways to handle vtables and by default gcc's version is not
what Windows is using.  The vtable-thunks flag is the "Windows" way?

3) And then there is the new gcc 3.0 coming up.  Am I correct in
understanding that the default implementation does the thunks that the old
vtable-thunks flag covered?  This means that I will not have to do anything
special to use OLE: no vtable-thunks flag, no special libs, etc?

TIA,

-pkg


_________________
Paul K Gleske
207.698.1742
AIM - pgleske
-----------------
"Make everything as simple as possible - but not simpler"  - A. Einstein