Quantcast
Channel: Parallel Computing in C++ and Native Code forum
Viewing all articles
Browse latest Browse all 856

Dynamically loading AMP kernels

$
0
0
Hi, I was wondering if you could help me with a problem I have. When using CUDA, I am able to build a kernel using NVIDIA's compiler, save it to a file (such as PTX or .cubin) and load it into the device using the CUDA Driver API.

My question: is something like this possible in C++ AMP? What I need to be able to do is generate C++ AMP kernels at run-time (as pieces of C++ code, though I'm open to suggestions), somehow compile then into independent units and then load them in. I understand that it's probably possible to build a DLL for each one of them, but I'm looking for something more direct.

- Dmitri Nesteruk, MVP C#, Microsoft Partner


Viewing all articles
Browse latest Browse all 856

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>