Description
use trtexec to convert the onnx to trt but failed. Error messages:
-------------- The current device memory allocations dump as below --------------
[0]:4194304 :HybridGlobWriter in reserveRegion: at optimizer/common/globWriter.cpp: 246 idx: 21185 time: 0.000145662
[0x21566a000]:6029312 :DeviceActivationSize in reserveNetworkTensorMemory: at optimizer/common/tactic/optimizer.cpp: 4596 idx: 4 time: 0.00196464
[08/08/2024-08:12:20] [W] [TRT] Requested amount of GPU memory (4194304 bytes) could not be allocated. There may not be enough free memory for allocation to succeed.
[08/08/2024-08:12:20] [W] [TRT] Skipping tactic 43 due to insuficient memory on requested size of 4194304 detected for tactic -1020632631321619146.
[08/08/2024-08:12:20] [V] [TRT] Fastest Tactic: -3360065831133338131 Time: inf
[08/08/2024-08:12:20] [E] Error[1]: [resizingAllocator.cpp::deallocate::100] Error Code 1: Cuda Runtime (misaligned address)
[08/08/2024-08:12:20] [V] [TRT] Deleting timing cache: 781 entries, 991 hits
[08/08/2024-08:12:20] [E] Error[10]: [optimizer.cpp::computeCosts::2011] Error Code 10: Internal Error (Could not find any implementation for node Conv_105 + Relu_106.)
[08/08/2024-08:12:20] [E] Error[2]: [builder.cpp::buildSerializedNetwork::609] Error Code 2: Internal Error (Assertion enginePtr != nullptr failed. )
[08/08/2024-08:12:20] [E] Engine could not be created from network
[08/08/2024-08:12:20] [E] Building engine failed
[08/08/2024-08:12:20] [E] Failed to create engine from model.
[08/08/2024-08:12:20] [E] Engine set up failed
Environment
Device: Jetson AGX Xavier
Jetpack Version: 4.6.2
TensorRT Version: 8.2.1
GPU Type:
Nvidia Driver Version:
CUDA Version:
CUDNN Version:
Operating System + Version:
Python Version (if applicable):
TensorFlow Version (if applicable):
PyTorch Version (if applicable): 1.9
Baremetal or Container (if container which image + tag):
Relevant Files
Steps To Reproduce
/usr/src/tensorrt/bin/trtexec --onnx=model.onnx --saveEngine=model.trt --fp16 --workspace=8192 --profilingVerbosity=detailed --verbose --dumpLayerInfo --minShapes=x:1x3x512x512 --optShapes=x:1x3x512x512 --maxShapes=x:1x3x512x512