Company Events Academic Community Support Solutions Products & Services Contact NI MyNI

Prepare Data For Lock VI

LabVIEW 2012 GPU Analysis Toolkit Help

Edition Date: June 2012

Part Number: 373575A-01

»View Product Info
Download Help (Windows Only)

Owning Palette: LVCUDA SDK

Requires: GPU Analysis Toolkit

Begins the process of locking device data.

Use this VI as the first step in protecting the device data from modification or deletion while in use. Wire data to the CUDA U8 Device Ptr in input to determine the polymorphic instance to use. You can wire a CUDA device pointer, CUDA context, CUFFT handle, or CUBLAS handle to this VI.

For more information about how to use the LVCUDA SDK and LVGPU SDK VIs, refer to the KnowledgeBase article Customizing GPU Computing Using the GPU Analysis Toolkit at ni.com.

The connector pane displays the default data types for this polymorphic function.

Details  

CUDA U8 Device Ptr in specifies a device pointer class. This input also accepts a CUDA context class, a CUFFT handle class, or a CUBLAS handle class. This input can contain the following data types:

  • 8-bit unsigned integers
  • 16-bit unsigned integers
  • 32-bit unsigned integers
  • 8-bit signed integers
  • 16-bit signed integers
  • 32-bit signed integers
  • Single-precision, floating-point numbers
  • Double-precision, floating-point numbers
  • Complex single-precision, floating-point numbers
  • Complex double-precision, floating-point
error in describes error conditions that occur before this node runs. This input provides standard error in functionality.
CUDA U8 Device Ptr out returns a device pointer class, CUDA context class, CUFFT handle class, or a CUBLAS handle class. This output returns the same data type that you wire to CUDA U8 Device Ptr in.
CUDA U8 Device Ptr Data Reference returns the device pointer, CUDA context, CUFFT handle, or CUBLAS handle reference to lock. This output returns a data value reference to a class with the same data type that you wire to CUDA U8 Device Ptr in.
CUDA Context returns the device context class in which the device pointer, the CUFFT handle, or the CUBLAS handle was created.
error out contains error information. This output provides standard error out functionality.

Prepare Data For Lock Details

The device data is stored using a data value reference returned in the CUDA Device Ptr out, CUDA Context out, CUFFT Handle out, or CUBLAS Handle out output. Wire this output to a Data Value Reference Read Element border node on an In Place Element structure, which locks the data value reference while you operate on the code inside the In Place Element structure. To unlock the device data, wire it to the Data Value Reference Write Element node. You can make the device data available to other processes by wiring the output of the Data Value Reference Write Element node to the Prepare Data for Unlock VI.


 

Your Feedback! poor Poor  |  Excellent excellent   Yes No
 Document Quality? 
 Answered Your Question? 
Add Comments 1 2 3 4 5 submit