DPC++ for NVIDIA/CUDA devices on Windows

[pclausen]

Hi,

I downloaded the oneAPI BasiKit and then oneAPI HPC on windows. I could run a sample code both from Visual Studio (2019) and also generated from cmake (3.19.2). The sample code compiles and runs. I have both an Intel Graphics 630 card and a NVIDIA Quadro M2200 card, but I can't see the NVIDIA device (see file attached); I also don't understand what the device [0x591b] refers to. Note that I can successfully run a CUDA examples in Visual Studio and also generated from cmake. From what I read, it seems that using DPC++ for CUDA devices is experimental and was tested on Linux only, but I am not sure. Is there a way to use it now on windows? If not, what is the time expectation to have it working on windows?

 

[GouthamK_Intel · Accepted solution]

Hi,

Thank you for your interest. We do not support DPC++ with CUDA BE on Windows at this point and have no plans for it in near future.

As you have already raised a similar issue in the below thread, we are closing this thread as it is a duplicate. we will no longer respond to this thread.
If you require any additional assistance from Intel, please start a new thread.
Any further interaction in this thread will be considered community only.

https://community.intel.com/t5/Intel-oneAPI-Data-Parallel-C/DPC-for-NVIDIA-CUDA-devices-on-Windows/m-p/1246200#M900

Have a Good day!

Thanks & Regards

Goutham