Initialize Device VI

Owning Palette: LVCUDA

Requires: GPU Analysis Toolkit

Creates the special execution environment, or context, necessary for LabVIEW to communicate with the GPU.

Details  Example

	CUDA Device ID specifies the device ID of the NVIDIA GPU. The default is 0.
	Select CUDA Runtime Version specifies the method you select to determine the version of CUDA to use at run time. 0 Default—On Windows, Default selects the CUDA version by examining the environment variable for the path containing the bin folder where the CUDA libraries are installed on the host computer. 1 By Environment—Selects the CUDA version by examining the environment variable for the path containing the bin folder where the CUDA libraries are installed on the host computer. 2 By Path—Selects the CUDA version by examining the path you provide in the CUDA Runtime Version Info input to where the CUDA libraries are installed.
	CUDA Runtime Version Info specifies additional information based on Select CUDA Runtime Version.
	error in describes error conditions that occur before this node runs. This input provides standard error in functionality.
	CUDA Context out returns the device context that manages device execution and device resources.
	CUDA Version returns the current CUDA version based on the method you select in the Select CUDA Runtime Version input.
	error out contains error information. This output provides standard error out functionality.

Initialize Device Details

For more information about this VI, refer to the Designing the Block Diagram to Compute on a GPU Device topic.

For more information about the CUDA API, refer to the NVIDIA GPU Computing Documentation website at nvidia.com and download the CUDA API Reference Manual.

Example

Refer to the Multi-channel FFT VI in the labview\examples\lvgpu\Multi-channel FFT directory for an example of using the Initialize Device VI.