Native Heterogeneous Computing for D
To use this package, put the following dependency into your project's dependencies section:
This project is a set of libraries designed to work with a modified ldc to enable native execution of D on GPUs (and other more exotic target of OpenCL, hereafter just 'GPUs').
There are three main parts
- A standard library 'std' containing standard functionality for targetting GPUs.
- A driver library to handle all the compute API interactions and provide a friendly easy-to-use consistent interface. Of course you can always get down to a lower level of interaction if you need to.
- A set of standard kernels to cover a large number of use cases and serve as documentation on how (and how not) to use this library.
To build DCompute you will need a modified ldc available here. In addition you will need Khronos's SPIRV-LLVM (available here) (use the spirv-3.8 branch) as well as a dmd compatible d compiler and cmake for building ldc.
Please see the wiki.
Find out what khronos's clang mangles the OpenCL builtins' names to so that we can use the correct ones.
Get D versions of the OpenCL and CUDA APIs so that work can begin on a unified D driver API.
- Registered by Ilya Yaroshenko
- Copyright © 2016, Nicholas Wilson
1 downloads today
2 downloads this week
2 downloads this month
12 downloads total