Nvidia works with LLVM developers to CUDA compiler to open source

Published:2012/5/10 20:01:00 Author:Ecco From:SeekIC

Nvidia Corp. announced a LLVM which is a popular open source compiler to support Nvida GPUs. According to Nvidia, they would provide the source code for its CUDA low-level virtual machine (LLVM)-based compiler to academic researchers and software-tool vendors last year.

Nvidia’s engineers work with LLVM developers to provide Nvidia’s CUDA compiler source code. And the code can be changed the LLVM core and parallel thread execution backend. Programmers can now develop applications for GPU accelerators using a broader selection of programming languages. The code Nvidia provided to LLVM is based on proven, mainstream CUDA products. It gives programmers the assurance of reliability and full compatibility with the hundreds of millions of Nvidia GPUs installed in PCs and servers today.

LLVM supports a range of programming languages and front ends, including Python, C/C++, Objective-C, Ada, Fortran, Haskell, Java bytecode, Ruby, ActionScript, GLSL and Rust. It is also the compiler infrastructure for its CUDA C/C++ architecture, and it has been widely adopted by leading companies such as Apple, AMD and Adobe. LLVM is a widely used open source compiler infrastructure with a modular design to make it easy to add support for programming languages and processor architectures.

Reprinted Url Of This Article: http://www.seekic.com/blog/IndustryNews/2012/05/10/Nvidia_works_with_LLVM_developers_to_CUDA_compiler_to_open_source.html

Print this Page | Comments | Reading(549)