I tried to launch sample_drivenet in my host PC when i had installed Driveworks SDK, but it showed this error: ./sample_drivenet: error while loading shared libraries: libcuda.so.1: cannot open shared object file: No such file or directory
I found libcuda.so.1 and add it’s address to the LD_LIBRARY_PATH like this: export LD_LIBRARY_PATH=/home/nvidia/nvidia_sdk/DRIVE/Linux/5.0.10.3/OS/DriveSDK/drive-t186ref-linux/lib-target:$LD_LIBRARY_PATH
but the error still remained, what can i do now?
Additionally,can i use driveworks SDK in VMware in my host PC?
I guess, you cannot run samples on VM. Atleast for me it didnt work. But I never tried on VMware. You need all the CUDA stuff installed on VMware to make sure it works.
And i tried to use google to find answers, some people said it’s wrong with my NVIDIA drive, so i changed my NVIDIA drive using NVIDIA-Linux-x86_64-396.18.run, and the error changed to:
[b][26-1-2019 5:37:55] CameraVirtual: The recorded RAW video appears to be a version 2, which is no longer supported
[26-1-2019 5:37:55] CameraRAW: no seek table found at …/./data/samples/raw/rccb.raw.seek, seeking is not available.
Camera image: 1920x1208
Camera image with 1920x1208 at 30 FPS
[26-1-2019 5:37:55] Driveworks exception thrown: DW_CUDA_ERROR: ImageStreamer(CUDA->GL): cannot register GL texture as resource : OS call failed or operation not supported on this OS
terminate called after throwing an instance of ‘std::runtime_error’
what(): [2019-01-26 05:37:55] DW Error DW_CUDA_ERROR executing DW function:
dwImageStreamer_initialize(&m_streamer, &imageProps, typeOut, ctx)
at …/samples/framework/…/framework/SimpleStreamer.hpp:85
Aborted (core dumped)[/b]
And when i changed the driver to NVIDIA 396.26, the error became: [26-1-2019 6:16:45] Sensor statistics for: camera.virtual, dla=0,dlaEngineNo=0,enableFoveal=0,precision=fp32,stopFrame=0,video=…/./data/samples/raw/rccb.raw
[26-1-2019 6:16:45] Events: 0
Errors: 0
Drops: 0
minDelta: 32389.0000000
maxDelta: 67632.0000000
meanDelta: 41253.617188
Standard deviation: 14336.022461
[26-1-2019 6:16:45] CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_medium_nn_v1.
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_medium_nn_v1.
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_medium_nn_v1.
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_medium_nn_v1.
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_medium_nn_v1.
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
[26-1-2019 6:16:45] caskConvolutionLayer.cpp (195) - Cuda Error in execute: 48
You can remove cuda and driveworks. Reinstall Driveworks, and it automatically install CUDA for you. For me DW0.6 works fine and you can DriveInstall 5.0.5.0b Linux version from the archive.