Thanks! I’m wondering about operations launched using the cuDNN interface. According to https://forums.developer.nvidia.com/t/does-cudnncreate-call-create-multiple-streams-internally/157422, “cuDNN can use internal streams for work.” How do I ensure that those internal streams are running with high priority?
friedj
3
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Cuda stream priorities | 0 | 55 | July 17, 2025 | |
| Priority of concurrent CUDA kernel execution on TX1 | 5 | 1457 | October 18, 2021 | |
| CUDA stream priority across processes(Windows OS) | 1 | 794 | May 31, 2019 | |
| CUDA Stream priorities, WDDM and TCC behavior | 0 | 1121 | June 8, 2021 | |
| Questions of CUDA stream priority | 10 | 4750 | April 19, 2023 | |
| How high priority stream preemption | 12 | 7333 | November 30, 2022 | |
| My GPU cannot support cuda stream priority | 4 | 392 | June 3, 2024 | |
| memcpy priority of cuda streams | 0 | 770 | March 9, 2020 | |
| Is there a way to allocate priority among different ExecutionContexts in TensorRT? | 1 | 1265 | December 8, 2020 | |
| cuda stream high priority could not always schedule high prority | 2 | 815 | July 11, 2019 |