tra added a comment. I'm not convinced that nvcc's behavior is a good guide for clang's defaults.
Considering that clang is not compatible with nvcc when it comes to command line options, whoever is using clang to compile CUDA already has to adjust command line options. Explicitly adding `-O3` during the process is, IMO, better than requiring everyone to be aware that if the input file is CUDA (or HIP), then clang would default to -O3. https://reviews.llvm.org/D54183 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
