Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 25 Jan 2013 22:51:18 +0100
From:      Dimitry Andric <dim@FreeBSD.org>
To:        Pedro Giffuni <pfg@FreeBSD.org>, Konstantin Belousov <kostikbel@gmail.com>
Cc:        freebsd-toolchain@freebsd.org
Subject:   Re: Removing default build of gcc
Message-ID:  <5102FE56.40806@FreeBSD.org>
In-Reply-To: <5102F107.8090501@FreeBSD.org>
References:  <74D8E686-3679-46F2-8A08-4CF5DFC020CA@FreeBSD.org> <20130125113122.GN2522@kib.kiev.ua> <E0EA1F1F-99BB-47F5-94A3-1C197F680BD9@bsdimp.com> <20130125195941.GW2522@kib.kiev.ua> <5102ECBF.4060500@FreeBSD.org> <20130125204430.GX2522@kib.kiev.ua> <5102F107.8090501@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On 2013-01-25 21:54, Pedro Giffuni wrote:
...
> I am aware a fix is being worked on. I think that as long as
> the default compiler/C++ library works it is OK to make things
> easier for other compilers. I am OK with having that change in
> -current but for 9.x it is simply unacceptable.

Actually, clang with libc++ works fine, and both clang and gcc with
libstdc++ don't...

If the problem is caused by the switchable libsupc++.so backend lib, I
would have no trouble with reverting that.  But do we know that for sure
at this point?  I have not spent enough time looking deeply into the
issue.

I did notice that libsupc++ .So objects get linked into libstdc++.so,
even while they are also in libsupc++.so.  Maybe that is causing trouble?



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5102FE56.40806>