User Guide

  • 2022.2
  • 04/11/2022
  • Public Content

Run a Group of Checks

The Diagnostics Utility for Intel® oneAPI Toolkits organizes checks in groups or categories. To choose specific groups you want to run, use the List of Checks.
For example, use the
argument to run multiple checks related to GPU hardware and software. The
argument will output these details:
  • OpenCL and LevelZero driver details
  • Identify if VTune
    Analyzer is ready to do analysis on GPUs
  • Identify Intel GPUs on this system
  • State of hangcheck option: enabled or disabled
  • Identify if the current user is in the same group as the GPU(s)
  • GPU metrics
  • Identify if GPU is ready to run applications
  • Identify if environment is ready to analyze GPU kernels
To run the utility with the
python3 --filter gpu
You will see a result similar to this:
Output showing all gpu checks
The information in the output indicates that a GPU was present, but does not provide any more information about the GPU or how many GPUs are present. To get more information, use
to activate verbose mode:
python3 --filter gpu -v
The output will contain expanded details for each check. In the example below, the
shows what information the Diagnostics Utility for Intel® oneAPI Toolkits found.
Output showing all gpu checks with verbosity
Verbosity can be customized to output different levels of details. To learn more, see Verbose Mode Options.
To see a full list of available groups, enter
python3 --list
in your terminal, or see List of Checks by Group Name.

Product and Performance Information


Performance varies by use, configuration and other factors. Learn more at