Fix memory consumption by PyTorch in dlpack zero-copy perf test.

[mzient]

Category:

Other (e.g. Documentation, Tests, Configuration)

Description:

The test used to create a separate CUDA/torch stream for each run, which resulted in unexpectedly high memory consumption in PyTorch. Using a single global stream fixes the issue.

Additional information:

Affected modules and functionalities:

Key points relevant for the review:

Tests:

• Existing tests apply
• New tests added
  • Python tests
  • GTests
  • Benchmark
  • Other
• N/A

Checklist

Documentation

• Existing documentation applies
• Documentation updated
  • Docstring
  • Doxygen
  • RST
  • Jupyter
  • Other
• N/A

DALI team only

Requirements

• Implements new requirements
• Affects existing requirements
• N/A

REQ IDs: N/A

JIRA TASK: N/A

[dali-automaton]

CI MESSAGE: [27382426]: BUILD STARTED

[JanuszL]

Just remove global as we don't assign anything to the variables in local scope. Other than that 👍

[dali-automaton]

CI MESSAGE: [27382555]: BUILD STARTED

[dali-automaton]

CI MESSAGE: [27382555]: BUILD FAILED

[dali-automaton]

CI MESSAGE: [27382555]: BUILD PASSED