Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 25 Apr 2015 16:45:14 +0200
From:      John Marino <freebsd.contact@marino.st>
To:        Don Lewis <truckman@FreeBSD.org>, jbeich@FreeBSD.org,  ports-committers@FreeBSD.org, svn-ports-all@FreeBSD.org,  svn-ports-head@FreeBSD.org
Subject:   Re: svn commit: r384676 - head/cad/kicad
Message-ID:  <553BA87A.4060002@marino.st>
In-Reply-To: <20150425143648.GA91290@graf.pompo.net>
References:  <618l-7yth-wny@FreeBSD.org> <201504250611.t3P6B37U069425@gw.catspoiler.org> <20150425143648.GA91290@graf.pompo.net>

next in thread | previous in thread | raw e-mail | index | archive | help

On 4/25/2015 16:36, Thierry Thomas wrote:
> Le sam 25 avr 15 à  8:11:03 +0200, Don Lewis <truckman@FreeBSD.org>
>  écrivait :
> 
>> I also noticed that brokenness in the Makefile, but I think libc++ isn't
>> getting linked in because of the presence of -nostdinc++ and the lack of
>> -lc++.  I think the three lines above should go away.
>>
>> The linker error seems to indicate that the linker is trying to link
>> against libstdc++ in base instead of the one bundled with the gcc port.
>> The likely cause is that the port is ignoring LDFLAGS, which contains
>> 	-Wl,-rpath=${_GCC_RUNTIME} -L${_GCC_RUNTIME}
>> because of USE_GCC=4.7.
> 
> When the default Gcc was upgraded from 4.7 to 4.8, all concerned ports
> were OK, but this one failed!
> 
> This is a work-around to keep it usable. Patches are welcome if you find
> a better solution!
> 
> Note: there are different ways to build it successfully (e.g.
> compiler:c++11-lib is sufficient), but then it does not run correctly.
> 

This is probably an ignorant question, but what does cad/kicad bring
over cad/kicad-devel, which is currently building normally IIUC.  Is
there concrete justification to have both at this point?  Maybe we can
just retire kicad.  It sounds fixable but it sounds like it's a lot of
work to fix too.

Just wondering,
John







Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?553BA87A.4060002>