search

autumnai / collenchyma

Extendable HPC-Framework for CUDA, OpenCL and common CPU
http://autumnai.github.io/collenchyma
Apache License 2.0
476 stars 33 forks source link

Extract CUDA wrapper #4

Open MichaelHirn opened 8 years ago

MichaelHirn commented 8 years ago

For Cuda we can take a lot of the structure alreay implemented for OpenCL. I also think we should go with implementing the lower-level Cuda Driver API instead of the Cuda Runtime API as it gives us more flexibility and is closer to the OpenCL behavior. Both implementations could be mixed, though.

Thanks to Bindgen, the extraction of the ffi becomes a way easier.

hobofan commented 8 years ago

Quite some time has passed since the issue was opened and a significant (usable) part has been implemented.

There are some benefits to having a seperate wrapper crate for CUDA, so that part should be extracted in the future.