Intel® oneAPI Data Parallel C++
Support for Intel® oneAPI DPC++ Compiler, Intel® oneAPI DPC++ Library, Intel® DPC++ Compatibility Tool, and GDB*
Announcements
The Intel sign-in experience has changed to support enhanced security controls. If you sign in, click here for more information.
443 Discussions

Use nVidia GPU via OpenCL with oneAPI on Windows

PatrickBenomard
1,471 Views

Hi,

Is there a way to use oneAPI to target nVidia GPUs via OpenCL on Windows?

I followed the vector-add and iso2dfd tutorials within Visual Studio 2019 but only the Intel GPU is detected.

Kind regards,

Patrick

0 Kudos
1 Solution
PatrickBenomard
1,162 Views

Hi Noorjahan,

 

Yes, you can close the thread, thanks.

 

Cheers,

 

Patrick

View solution in original post

9 Replies
NoorjahanSk_Intel
Moderator
1,423 Views

Hi,


Thanks for reaching out to us.


>>Is there a way to use oneAPI to target nVidia GPUs via OpenCL on Windows?


As you wanted to use NVIDIA GPU, you can download opensource oneAPI to target on non-intel GPU's, as you cannot use Intel oneAPI to target NVIDIA GPUs via OpenCL


You can refer to the below link to download opensource oneAPI:


https://www.oneapi.io/open-source/


>> followed the vector-add and iso2dfd tutorials 


You can try the code from below link to target Nvidia GPU.

https://developer.codeplay.com/products/computecpp/ce/guides/sycl-for-cuda-developers/examples


Thanks & Regards

Noorjahan.



NoorjahanSk_Intel
Moderator
1,295 Views

Hi,


Has the information provided helped? If yes could you please confirm whether can we close this thread from our end?


Thanks & Regards,

Noorjahan.


PatrickBenomard
1,278 Views

Hi,

 

I only had a quick look at your answer for now, and am not sure whether it is going to help.

The below code does not print any other GPU device than the "Intel(R) HD Graphics 630" one on my machine, whereas it has a nVidia one too.

Is there a way to have GPUs other than Intel ones be "seen"?

 

for (auto device : sycl::device::get_devices(sycl::info::device_type::gpu))
{
    std::cout << " Device: " << device.get_info<sycl::info::device::name>() << std::endl;
}

 

Kind regards,

 

Patrick

NoorjahanSk_Intel
Moderator
1,260 Views

Hi,


Could you please let us know from where you have installed Intel oneAPI Toolkits?


>> does not print any other GPU device than the "Intel(R) HD Graphics 630"


It will not print any other device if you are using Intel oneAPI. If you want to target the NVidia GPU, you can use open-source oneAPI.


You can refer to the below links for more details:

https://www.oneapi.io/open-source

https://intel.github.io/llvm-docs/GetStartedGuide.html


Thanks & Regards

Noorjahan.



PatrickBenomard
1,214 Views

Hello,

 

I indeed got oneAPI from https://www.intel.com/content/www/us/en/developer/tools/oneapi/base-toolkit-download.html, which is Intel version.

I will give the open source version a shot.

 

Shall I uninstall Intel oneAPI toolkits first?

 

Kind regards,

 

Patrick

NoorjahanSk_Intel
Moderator
1,189 Views

Hi,


>>Shall I uninstall Intel oneAPI toolkits first?


You can make use of open-source oneAPI even without uninstalling Intel OneAPI toolkits.


Please follow the below link for more details regarding the installation of opensource oneAPI:

https://intel.github.io/llvm-docs/GetStartedGuide.html


Thanks & Regards,

Noorjahan.


NoorjahanSk_Intel
Moderator
1,161 Views

Hi,

Has the information provided above helped? If yes, could you please confirm whether we can close this thread from our end?


Thanks & Regards

Noorjahan.




PatrickBenomard
1,163 Views

Hi Noorjahan,

 

Yes, you can close the thread, thanks.

 

Cheers,

 

Patrick

NoorjahanSk_Intel
Moderator
1,120 Views

Hi,

Thanks for the confirmation. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.


Thanks & Regards,

Noorjahan.


Reply