On Feb 6, 2013, at 1:56 PM, Roy Stogner <royst...@ices.utexas.edu> wrote:
>> >> What I am trying to accomplish is to force netcdf to use our >> compilers. Without this it''l happily go and find its own, >> regardless of what we are using. note that this is not strictly a >> problem but seems silly to me. > > Oh, that would have been a horrible problem. If distcc fails to > configure, that's still hugely preferable to distcc getting silently > unused in part of a "make -j 40" on a machine with a finite amount of > RAM. There may be other ways do accomplish this (export CXX from whatever we set it to?) but this seemed the most clear. I just thought about the case where the user passes libMesh CXX, and while redundant this still seems to work: $ ../configure CXX=clang++ … === configuring in contrib/netcdf/v4 (/Users/benkirk/codes/libmesh.benkirk/clang/contrib/netcdf/v4) configure: running /bin/sh ../../../../contrib/netcdf/v4/configure --disable-option-checking '--prefix=/usr/local' 'CXX=clang++' 'CC=clang' 'libmesh_CXXFLAGS=-DOMPI_SKIP_MPICXX' 'PETSC_DIR=/usr/local/petsc/3.3-p5' 'PETSC_ARCH=macosx' 'SLEPC_DIR=/usr/local/slepc/3.3-p3' 'TRILINOS_DIR=/usr/local/trilinos/11.0.3' '--disable-netcdf-4' 'CXX=clang++' 'CC=clang' 'F77=mpif77' 'FC=mpif90' --cache-file=/dev/null --srcdir=../../../../contrib/netcdf/v4 So CXX actually gets passed twice, but at least it is guaranteed consistent. -Ben ------------------------------------------------------------------------------ Free Next-Gen Firewall Hardware Offer Buy your Sophos next-gen firewall before the end March 2013 and get the hardware for free! Learn more. http://p.sf.net/sfu/sophos-d2d-feb _______________________________________________ Libmesh-devel mailing list Libmesh-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-devel
