This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TDA4VM: TDA4 GPU usage scenario

Part Number: TDA4VM

Hi,

    Currently, we know that the GPU is only available on A72 Linux via OpenGL. Is there any way to implement the algorithm on DSP that can also be used on GPU?

    Thank you !

    Regards,

    Kong

  • Hi,

         Also, we hope you can provide some reference cases and tutorials on specific uses of the GPU via OpenGL on A72 Linux.

          At the same time, if there are some ways in which the algorithm implementation on the DSP can use the GPU?

          In addition to the above, if you can use GPU in other ways, I hope you can provide some relevant case tutorials and references.

          Thanks!

         Regards,

          Kong

  • Hello Kong,

    We currently do not have support for offloading the DSP onto the GPU. As you said, we only support OpenGL on the GPU through it's Linux driver.

    What you are mentioning is something like OpenCL support where we can run algorithms between cores as compute, but this is not available in our SDK offering.

    You can run data through the DSP, then pass-along that data to the GPU for processing using OpenGL. But this framework is not in place currently.

    Regards,

    Erick