Mailing-List: contact cygwin-help@sourceware.cygnus.com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner@sources.redhat.com Delivered-To: mailing list cygwin@sources.redhat.com X-Authentication-Warning: hp2.xraylith.wisc.edu: khan owned process doing -bs Date: Fri, 30 Mar 2001 00:59:57 -0600 (CST) From: Mumit Khan To: cygwin@cygwin.com Subject: Re: -mno-cygwin and C++ -- solved! In-Reply-To: <20010329234651.A10894@redhat.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 29 Mar 2001, Christopher Faylor wrote: > > Wasn't --mno-cygwin a Geoff Noer invention? My memory rusty on this, but I believe you're correct (Steve Chamberlain would've been my second guess, but he may have been before Colin Peters' work in developing mingw). > Would I cause a massive volcanic eruption if I ripped --mno-cygwin out > of gcc and just suggested that people ran the mingw version of gcc? > I guess we'd then be subjected to pathname complaints... As you note in other messages, Cygwin itself needs the -mno-cygwin. My point is that if you want to use mingw with a unix style configuration scheme, you need to really build it as a cross-compiler to avoid the various pitfalls in using -mno-cygwin. I still have a build system from the old days that basically renamed i686-pc-mingw-gcc to just gcc, had a uname that spit out mingw in some form (Earnie's contribution a long time ago), and a few other tweaks, and I could build my own packages which are rather large. Of course, I'd mostly build on either linux or solaris simply because hosting quality, but that's not much of an issue anymore with Cygwin's dramatic improvement of late, especially under w2k. Regards, Mumit -- Want to unsubscribe from this list? Check out: http://cygwin.com/ml/#unsubscribe-simple