AnsweredAssumed Answered

Directly using OpenCL LLVM Compiler Backend

Question asked by keldor314 on Jun 9, 2012
Latest reply on Jun 11, 2012 by MicahVillmow

Is there any way of bypassing the C frontend of the OpenCL compiler and compiling directly from LLVM IR?  This would give us a practical means of writing our own compilers...

 

Nvidia offers this capability through their NVVM compiler SDK as of the CUDA 5.0 preview, and it obviously works on CPUs, so AMD GPUs would be the odd one out.

Outcomes