# Notice

This task is an original synthetic fixture. No upstream source code was copied.

It is inspired by public CUDA stream/event ordering guidance and public cuDF
native/JVM wrapper concepts. The starter program is intentionally small so the
task focuses on object readiness, cross-stream consumption, and device-memory
lifetime ordering.