I'm fairly confident this isn't a compiler problem, but a libc++ problem.
CXXFLAGS="-stdlib=libstdc++" cmake ..
gets you past the current problem. But if boost is built against libc++ ,
it doesn't link, so I also built boost against libstdc++ and everything
built.
Based on Greg's playing with emscripten a few months back [1], an update to
libc++ may fix everything. Then again, I'm not sure whether apple would
libc++ with minor os releases.
[1]
https://groups.google.com/d/msg/emscripten-discuss/WyERe1VDl94/QmSRbe6tY3sJ
-David
On Sat, Oct 26, 2013 at 12:15 AM, Greg Landrum <[email protected]>wrote:
>
>
> On Fri, Oct 25, 2013 at 11:06 PM, William G. Scott <[email protected]>wrote:
>
>> Hi Greg:
>>
>> I’ve just placed two log files on
>> http://fennario.ucsc.edu/~wgscott/temp/rdkit/
>>
>
> The error messages look really familiar, but I unfortunately can't find a
> reference for them. I will keep dredging around though.
>
> One was generated using the /usr/bin/g++ compiler on 10.9, i.e.,
>>
>> zsh-% /usr/bin/g++ --version
>> Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr
>> --with-gxx-include-dir=/usr/include/c++/4.2.1
>> Apple LLVM version 5.0 (clang-500.2.79) (based on LLVM 3.3svn)
>> Target: x86_64-apple-darwin13.0.0
>> Thread model: posix
>>
>> which seems to be the same as
>>
>> zsh-% /usr/bin/clang++ --version
>> Apple LLVM version 5.0 (clang-500.2.79) (based on LLVM 3.3svn)
>> Target: x86_64-apple-darwin13.0.0
>> Thread model: posix
>>
>>
>> The other one, which gets a bit further, was generated with fink’s g++
>> version 4.8 compiler in /sw/bin/g++, i.e.,
>>
>> zsh-% /sw/bin/g++-4 --version
>> g++-4 (GCC) 4.8.2
>> Copyright (C) 2013 Free Software Foundation, Inc.
>>
>>
>> However, as a positive control, I just installed the most recently
>> available compilers for 10.8, which still works fine to compile rdkit
>>
>> fennario-% /usr/bin/g++ --version
>> Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr
>> --with-gxx-include-dir=/usr/include/c++/4.2.1
>> Apple LLVM version 5.0 (clang-500.2.76) (based on LLVM 3.3svn)
>> Target: x86_64-apple-darwin12.5.0
>> Thread model: posix
>>
>> It looks almost the same as the one for 10.9, so I am even more stumped
>> than before.
>
>
> Any chance you could try on the 10.9 system with either a different
> version of boost (1.51?) or the new RDKit beta?
> We've got some evidence that it's not the compiler (I also did a clang3.3
> build on my linux box without problems), so trying an alternate boost
> version seems like the next logical step.
>
> Sorry for the inconvenience,
> -greg
>
>
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most
> from
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk
> _______________________________________________
> Rdkit-discuss mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>
>
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135991&iu=/4140/ostg.clktrk
_______________________________________________
Rdkit-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
