Mailing-List: contact cygwin-apps-help AT cygwin DOT com; run by ezmlm
Sender: cygwin-apps-owner AT cygwin DOT com
List-Subscribe: <mailto:cygwin-apps-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin-apps/>
List-Post: <mailto:cygwin-apps AT cygwin DOT com>
List-Help: <mailto:cygwin-apps-help AT cygwin DOT com>, <http://sources.redhat.com/lists.html#faqs>
Delivered-To: mailing list cygwin-apps AT cygwin DOT com
Message-ID: <0a2f01c1a073$b9f26bc0$0200a8c0@lifelesswks>
From: "Robert Collins" <robert DOT collins AT itdomain DOT com DOT au>
To: <cygwin-apps AT cygwin DOT com>
References: <20020118180305 DOT GA2312 AT dothill DOT com> <20020118180632 DOT GA24778 AT redhat DOT com>
Subject: Re: gcc -mno-cygwin STL support for setup.exe?
Date: Sat, 19 Jan 2002 09:59:03 +1100
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
X-OriginalArrivalTime: 18 Jan 2002 23:01:03.0122 (UTC) FILETIME=[00AF3F20:01C1A074]


===
----- Original Message -----
From: "Christopher Faylor" <cgf AT redhat DOT com>

> Weren't you going to rewrite rebase in C?  It is very likely that it
wouldn't
> be accepted into binutils otherwise.
> >About a year ago, there was some discussion about adding libstdc++.a
to
> >Cygwin's gcc -mno-cygwin mode.  Can we revisit this issue?
>
> The biggest issue with relying on libstdc++.a for for setup.exe is
that it
> requires this library to be built for cross-compilation.  I have no
idea how
> to do that.
>
> We have a looming problem here, though, since we will need to include
at
> least a libsupc++.a for -mno-cygwin if/when we start using gcc 3.x.  I
can't
> build setup.exe with newer versions of gcc currently due to this lack.
> I don't know if libstdc++.a even builds under mingw currently.

I presume you mean mingw with gcc 3?

This could well be a serious issue as we use new() currently. I hate to
suggest it, but maybe having a contributed cross-compile package really
is the way to go. Sure, it'll massively increase
build-cross-compile-from-scratch overhead (build cygwin target, build
mingw target, build cygwin hosted) ....

I suspect that we'll need a mingw target cross compiler to generate the
correct threading model for libstc++ at a minimum.

Rob