Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 03 Nov 2013 15:33:54 +0100
From:      John Marino <freebsd.contact@marino.st>
To:        Gerald Pfeifer <gerald@pfeifer.com>
Cc:        svn-ports-head@freebsd.org, Thierry Thomas <thierry@FreeBSD.org>, svn-ports-all@freebsd.org, Gerald Pfeifer <gerald@FreeBSD.org>, ports-committers@freebsd.org
Subject:   Re: svn commit: r332557 - head/lang/gcc
Message-ID:  <52765ED2.8080205@marino.st>
In-Reply-To: <alpine.LNX.2.00.1311031456490.3029@tuna.site>
References:  <201311022320.rA2NKEcg089316@svn.freebsd.org> <20131103110034.GA80884@graf.pompo.net> <alpine.LNX.2.00.1311031456490.3029@tuna.site>

next in thread | previous in thread | raw e-mail | index | archive | help
On 11/3/2013 14:59, Gerald Pfeifer wrote:
> On Sun, 3 Nov 2013, Thierry Thomas wrote:
>>>   Update to GCC 4.6.4 which, among others, addresses document building
>>>   with new versions of texinfo. [1]
>> Do we still need lang/gcc and lang/gcc46?
> 
> Yes, since lang/gcc is the canonical version that should be used
> (and is pulled in by USE_GCC=yes, for example).  At that point
> lang/gcc46 will be the final release of GCC 4.6, lang/gcc47
> will track GCC 4.7 and lang/gcc will track GCC 4.7 releases.
> 
> lang/gcc should have been upgraded to GCC 4.7 already (and hopefully
> will in the not too far future, there are just still a number of broken 
> ports -- see below for those not Python-related).
> 
> Gerald
> 
> 
> Done. List of failures and logs:
> 
> http://package20.nyi.freebsd.org/bulk/91amd64-default-gccupdate/2013-10-22_01h02m27s/
> 
>> + {"origin"=>"biology/plink", "pkgname"=>"plink-1.07", "phase"=>"build", "errortype"=>"cluster"}
> 
> C++ error. [snipped]

DragonFly has moved to use gcc 4.7 already and may already have the
fixes.  For example, plink builds on gcc 4.7 in DragonFly right now.

So for sure some of these can be transferred to avoid rework.

However, food for thought: Moving to gcc47 and exposing the resulting
package could have the benefit of others providing the fixes.  I can
tess you from experience these last 3 weeks that c++ breakage is very
slow to fix so people will be waiting a long time.  I've been doing this
for all the 4.4 to 4.7 fallout I've seen.  Luckily much of those fixes
I've put directly in ports.

John



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?52765ED2.8080205>