jhuber6 wrote: It's passed to the linker wrapper because it needs `fatbinary` for CUDA, but it should *also* be forwarded to the embedded clang job. If you compile with `-v` do you not see it on the `clang --target=nvptx64-nvidia-cuda` job?
https://github.com/llvm/llvm-project/pull/149107 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
