Date: Fri, 21 Dec 2012 19:01:52 -1000 (TAHT) From: Gerald Pfeifer <gerald@pfeifer.com> To: Andriy Gapon <avg@FreeBSD.org> Cc: svn-ports-head@FreeBSD.org, svn-ports-all@FreeBSD.org, Gerald Pfeifer <gerald@FreeBSD.org>, ports-committers@FreeBSD.org Subject: Re: svn commit: r307045 - head/Mk Message-ID: <alpine.LNX.2.00.1212211859090.2358@tuna.site> In-Reply-To: <5098E619.3070902@FreeBSD.org> References: <201211060023.qA60NhFW028290@svn.freebsd.org> <5098E619.3070902@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 6 Nov 2012, Andriy Gapon wrote: >> This extends revision r246991 (2010-01-02) and should not be necessary >> in most cases since LDFLAGS already covers linking, but one can always > > Rather than adding this flag to CXXFLAGS why not drop it from CFLAGS? > There is no place for linker options in compilation flags. Yes, but, there is an amazing lot of broken software out there. And, unlike compilation errors, failure to locate run-time libraries (or the right ones) is a lot harder to detect and we currently don't have a way to do so automatically. >> compile and link in one swoop, and this makes things consistent between >> C and C++. > This is a strange argument. When one does compilation and linking in > one swoop one uses both CFLAGS/CXXFLAGS and LDFLAGS. In non-broken > software there is never a dependency on linker flags auto-magically > appearing in CFLAGS/CXXFLAGS. If we'd only be dealing with non-broken software (or well maintained ports), I could have saved a couple of days of FreeBSD work this past year alone. :-/ I'm not opposed to removing this from both CFLAGS and CXXFLAGS, it "just" needs committment by all port maintainers to fix, if not test, their ports accordingly. That's the crux I see. Gerald
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.LNX.2.00.1212211859090.2358>