mstar.utils.streams#
CUDA stream state a worker has to manage.
What runs on a side stream matters: work on the SMs there (a kernel, or a device-to-device copy) while a CUDA graph replays slows every graph node in the process from then on. Side-stream work beside a live graph should be H2D / D2H copies only – see mstar.utils.h2d.
Functions
|
Create and destroy a throwaway CUDA context on |
- mstar.utils.streams.reset_device_scheduling(device)[source]#
Create and destroy a throwaway CUDA context on
device.Side-stream kernels or D2D copies beside a graph replay can leave graph nodes launching late until the device’s scheduling is rebuilt, which another context arriving on the GPU does.
- Return type:
None