Parallel Thread Execution

<p>Parallel Thread Execution (PTX or NVPTX) is a low-level <a href="/facts/Parallel_computing/nQzcpzQt">parallel</a> <a href="/facts/Thread_(computing)/hh8WHyPd">thread</a> <a href="/facts/Execution_(computing)/1fyfGMFm">execution</a> <a href="/facts/Virtual_machine/BLevU3sx">virtual machine</a> and <a href="/facts/Instruction_set_architecture/8p8vwUeE">instruction set architecture</a> used in <a href="/facts/Nvidia/8QfCgqbu">Nvidia</a>'s Compute Unified Device Architecture (<a href="/facts/CUDA/fvPpolot">CUDA</a>) programming environment. The <a href="/facts/Nvidia_CUDA_Compiler/JoU0JCsz">Nvidia CUDA Compiler</a> (NVCC) translates code written in CUDA, a <a href="/facts/C%252B%252B/Et0F2qn9">C++</a>-like language, into PTX instructions (an <a href="/facts/Intermediate_language/s0qRrbfh">IL</a>), and the graphics driver contains a <a href="/facts/Compiler/WNfCDFJe">compiler</a> which translates PTX instructions into executable binary code, which can run on the processing cores of <a href="/facts/List_of_Nvidia_graphics_processing_units/AXdfrvKo">Nvidia</a> <a href="/facts/Graphics_processing_unit/PTK1RQVp">graphics processing units</a> (GPUs). The <a href="/facts/GNU_Compiler_Collection/t07jh01k">GNU Compiler Collection</a>  and <a href="/facts/LLVM/nPmcspQN">LLVM</a> also have the ability to generate PTX. Inline PTX assembly can be used in CUDA.
</p>

Parallel Thread Execution open-in-new

Parallel Thread Execution