Hi charles
I have some answers (I hope this complete but there have been several attempts
so I would have to do it all over again from a clean sheet to prove it,
possibly not one for today)
here is what I did to get a good compile (this was under home-brew)
1. install gfortran from homebrew using there instructions, I am using
GNU Fortran (GCC) 4.8.2
2. add to arch/equivList
3. make the directory Darwin_13_x86_64
4. cp -r Darwin_12_x86_64 Darwin_13_x86_64
5. in Makefile.loc add
LOCAL_F77=gfortran
6. to avoid picking up the Accelerate framework export
LIBRARY_PATH=/usr/local//Cellar/gfortran/4.8.2/gfortran/lib:/Users/garythompson/programs/xplor-nih/2.35.0/bin.Darwin_13_x86_64
7. make some patches to satisfy clang
ensemblePot line 39 ./vmd/tclStream.cc TCLStream::operator void*() {return *in;}
to TCLStream::operator void*() {return in;}
line 115 /Users/garythompson/programs/xplor-nih/2.35.0/common/ensemblePot.hh
rc_DerivedEnsemblePot(Pot* realPtr = 0) : rc_EnsemblePot(realPtr) {}
rc_DerivedEnsemblePot(const rc_DerivedEnsemblePot& rhs) : rc_EnsemblePot(rhs)
rc_Pot -> rc_EnsemblePot
line 131 /Users/garythompson/programs/xplor-nih/2.35.0/CDSlib/array3D.cc
for (int i=0 ; i<size1()*size2()*size3 ; i++) ->
for (int i=0 ; i<size1()*size2()*size3() ; i++)
this compiles and produces and xplor executable command line there could be
more pain to come when I test it but we will have to see ;-)
regards
gary
nb I think the comment below would be absolutely true if i linked to lib
accelerate rather than to the inbuilt boas etc from fortlib
On 18 Mar 2014, at 19:50, Charles Schwieters <char...@schwieters.org> wrote:
>
> Hello Gary--
>
>>
>> from this it looks as if i use the accelerate framework I need to
>> use ―ff2c as well ??
>
> I don't think so, but I have never compiled using clang...
>
> best regards--
> Charles
_______________________________________________
Xplor-nih mailing list
Xplor-nih@cake.cit.nih.gov
http://cake.cit.nih.gov/mailman/listinfo/xplor-nih
