Date: Thu, 07 Feb 2013 10:45:31 -0500 From: Pedro Giffuni <pfg@FreeBSD.org> To: Baptiste Daroussin <bapt@FreeBSD.org> Cc: freebsd-office@FreeBSD.org Subject: Re: Problem with Boost and long double Message-ID: <5113CC1B.4000804@FreeBSD.org> In-Reply-To: <20130207071130.GC17793@ithaqua.etoilebsd.net> References: <51132FD0.4090304@FreeBSD.org> <20130207071130.GC17793@ithaqua.etoilebsd.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi Baptiste; On 02/07/2013 02:11, Baptiste Daroussin wrote: > On Wed, Feb 06, 2013 at 11:38:40PM -0500, Pedro Giffuni wrote: >> Hello; >> >> In some boost related code I am working on I found problems [1] thanks >> to the recent patch that enabled long double in boost port. >> >> >> For an explanation on why our libstdc++ doesn't support long double, >> you have to refer to this email: >> >> http://lists.freebsd.org/pipermail/freebsd-hackers/2009-March/028030.html >> >> Curiously it has just been discussed in -toolchain that our >> libstdc++ is misconfigured: >> >> http://docs.freebsd.org/cgi/mid.cgi?510EE5D2.4050409 >> >> Perhaps the patch in the post helps fix the issues but for now I would >> request that boost port disables again the long double support. >> >> Thanks, >> >> Pedro. >> >> [1] https://svn.boost.org/trac/boost/ticket/7987 > erg I have been happily building on all packages needing boost on my exp-run and > some people ensured me double long was working correctly now :( I guess that I am the only one trying to use Boost to calculate Bessel functions ;). I can send you a patch for openoffice if you want to reproduce it. > But sure I will disable boost as soon as I can, I need to rebuild every ports to > make sure everything is ok (mini exp-run) so won't happen before monday. The problem is libstdc++ so if libc++ is used (like in -current) there should be no problem. The configuration update may also fix things but that requires re-building world and all C++ ports. Pedro.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5113CC1B.4000804>