LLVM's Clang Lands More CUDA Improvements

Just days after writing about GPUCC as Google's open-source CUDA compiler built atop LLVM and how to compile CUDA code with LLVM, more improvements have landed.

There's now support for CUDA compilation by default as one of the most prominent changes today. "Currently clang requires several additional command line options in order to enable new features needed during CUDA compilation. This patch makes these options default." That change was done by Artem Belevich at Google.

Other work done by Artem today includes a new wrapper header for CUDA, detecting and linking CUDA's libdevice bitcode library, and other changes relating to better out-of-the-box CUDA support with LLVM/Clang.

This work will be found in the upcoming LLVM Clang 3.8 release.