b66pak
23rd December 2011, 19:25
On December 13th, NVIDIA announced that it will open up the CUDA platform by releasing source code for the CUDA Compiler.
Using the source code for new LLVM-based compiler that will be released with CUDA Toolkit 4.1, researchers and tools vendors will be able to:
* Support GPU acceleration in more programming languages, including many new and existing domain-specific languages (DSLs)
* Make new advancements in processor architecture and heterogeneous systems research
* Support CUDA applications on a wide variety of many-core and multi-core processors
source (http://developer.nvidia.com/content/cuda-platform-source-release)
_
Using the source code for new LLVM-based compiler that will be released with CUDA Toolkit 4.1, researchers and tools vendors will be able to:
* Support GPU acceleration in more programming languages, including many new and existing domain-specific languages (DSLs)
* Make new advancements in processor architecture and heterogeneous systems research
* Support CUDA applications on a wide variety of many-core and multi-core processors
source (http://developer.nvidia.com/content/cuda-platform-source-release)
_