search

NVIDIA / tensorrt-laboratory

Explore the Capabilities of the TensorRT Platform
https://developer.nvidia.com/tensorrt
BSD 3-Clause "New" or "Revised" License
261 stars 50 forks source link

run multiple models at one time on xavier #34

Closed jeansely closed 4 years ago

jeansely commented 4 years ago

Hi, how can i take two model at same time on nvidia jetson xavier. with last update i have error. [cuda.memcpy_htod_async(inp.device, inp.host, stream) for inp in inputs] pycuda._driver.LogicError: cuMemcpyHtoDAsync failed: invalid argument CException ignored in: <module 'threading' from '/usr/lib/python3.6/threading.py'> Traceback (most recent call last): File "/usr/lib/python3.6/threading.py", line 1294, in _shutdown t.join() File "/usr/lib/python3.6/threading.py", line 1056, in join self._wait_for_tstate_lock() File "/usr/lib/python3.6/threading.py", line 1072, in _wait_for_tstate_lock elif lock.acquire(block, timeout): KeyboardInterrupt FATAL: exception not rethrown Aborted (core dumped)

ryanolson commented 4 years ago

Jetson is outside the scope of this project. However, you can use the examples in this project to use TensorRT efficiently on Jetson. See Issue #35.

However, your example code above looks to be some issues with using pycuda.