X-Spam-Check-By: sourceware.org
Message-ID: <441BA2B3.C4CF018D@dessent.net>
Date: Fri, 17 Mar 2006 22:03:31 -0800
From: Brian Dessent <brian AT dessent DOT net>
MIME-Version: 1.0
To: cygwin AT cygwin DOT com
Subject: Re: GCC 4.x+
References: <Pine DOT OSF DOT 4 DOT 21 DOT 0603160045360 DOT 28991-100000 AT ax0rm1 DOT roma1 DOT infn DOT it> <4418AB38 DOT B7D87B36 AT dessent DOT net> <441B8AB3 DOT 9080308 AT cwilson DOT fastmail DOT fm>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-IsSubscribed: yes
Reply-To: cygwin AT cygwin DOT com
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

Charles Wilson wrote:

> I thought there were some patches to the cygwin gcc 3.4.x version that
> had not yet been migrated to the official sources?  I'd be glad to be
> wrong, however.

There's the patch that fixes throwing exceptions across DLLs (or
something like that.)  I know it's been submitted by Danny but I think
it withered and he wasn't interested in championing it.  The thing is,
4.x is so vastly different than 3.x that I'm not sure if it's still even
needed or not.

> Also, wasn't there some issue with the std::string implementation that
> was causing problems for both cygwin-special and mingw-special g++?

Yes, this can be fixed (with a somewhat severe performance penalty) by
configuring libstdc++ with --enable-fully-dynamic-string.  There's a
patch against 3.4 that fixes this in a less invasive way, but no
equivalent for 4.x AFAIK.  On this issue though even the current 3.4 gcc
packages are broken, so it wouldn't be a regression per se.

> Otherwise, if it's so simple, I don't understand why Gerrit hasn't
> released gcc-4.x as a test version, nor [OT:] why Danny hasn't released
> a gcc-4.0 candidate for mingw.

Well, I know there still some rough edges with 4.x and cygwin/mingw,
which might prevent the maintainers from deploying it just on the basis
that a somewhat-broken compiler is worse than no compiler.  The last
time I tried a cygwin1.dll compiled with gcc 4.x it was nowhere close to
being functional, presumably because the stricter enforcement of
aliasing rules or more aggressive optimizations uncovered code problems.

But the person that started the thread just wanted to try GCC on his own
code, which sounded to me like the appropriate thing is just build it
and find out, since it's trivial to keep a number of independant
installed versions of gcc.

Brian

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/