I’m trying to use clang++ instead of nvcc to build lulesh 2.0 cuda version. And it fails in the compilation with errors like the one below:
opt/common/cuda/cuda-7.5.18/include/thrust/iterator/iterator_adaptor.h:187:5: error: expected member name or ‘;’ after declaration specifiers
It looks like clang++ is complaining about the thrust library comes with cuda, while nvcc works fine for this case. So does it mean clang++ does not support compiling this benchmark (even though it’s cuda 7.5)? Or how should I tackle these errors?
Thanks a lot!