TensorRT OSS make error on Jetson AGX: cannot built target trtexec
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 13.4k
- Forks
- 2.4k
- Avg merge
- 5d 3h
- Merged PRs (30d)
- 2
Description
Description
TensorRT OSS make error on Jetson AGX: cannot built target trtexec . This is preventing from executing the make completely.
Environment
TensorRT Version: 8.0.1.6
NVIDIA GPU:
NVIDIA Driver Version:
CUDA Version: 10.2
CUDNN Version: 8.2.1.32
Operating System: 18.04
Python Version (if applicable):
Tensorflow Version (if applicable):
PyTorch Version (if applicable):
Baremetal or Container (if so, version):
Steps To Reproduce
- I am trying to build TensorRT OSS from this branch https://github.com/NVIDIA/TensorRT/tree/release/8.0 as I am using TensorRT version 8.0.1.6
- I tried to Native build on Jetson (aarch64) with cuda-10.2 using the command
cd $TRT_OSSPATH
mkdir -p build && cd build
cmake .. -DTRT_LIB_DIR=$TRT_LIBPATH -DTRT_OUT_DIR=`pwd`/out -DTRT_PLATFORM_ID=aarch64 -DCUDA_VERSION=10.2 -DGPU_ARCHS=72 -DCMAKE_C_COMPILER=/usr/bin/gcc
CC=/usr/bin/gcc make -j$(nproc)
- The error is generated when I execute this line
CC=/usr/bin/gcc make -j$(nproc)
ake[2]: Entering directory '/home/agrosy/git/TRT_docker/TensorRT/build'
[ 98%] Built target sample_onnx_mnist_coord_conv_ac
[ 98%] Built target sample_uff_plugin_v2_ext
[ 99%] Linking CXX executable ../../out/trtexec
cd /home/agrosy/git/TRT_docker/TensorRT/build/samples/trtexec && /usr/local/bin/cmake -E cmake_link_script CMakeFiles/trtexec.dir/link.txt --verbose=1
/usr/bin/g++ -Wno-deprecated-declarations -DBUILD_SYSTEM=cmake_oss -Wl,--exclude-libs,ALL CMakeFiles/trtexec.dir/__/common/sampleEngines.cpp.o CMakeFiles/trtexec.dir/__/common/sampleInference.cpp.o CMakeFiles/trtexec.dir/__/common/sampleOptions.cpp.o CMakeFiles/trtexec.dir/__/common/sampleReporting.cpp.o CMakeFiles/trtexec.dir/trtexec.cpp.o CMakeFiles/trtexec.dir/__/common/logger.cpp.o -o ../../out/trtexec /usr/local/cuda-10.2/lib64/libcudart.so /usr/local/cuda-10.2/lib64/libcublas.so /usr/lib/aarch64-linux-gnu/libcudnn.so /usr/lib/aarch64-linux-gnu/libnvinfer.so /usr/lib/aarch64-linux-gnu/librt.so -ldl -lpthread /usr/lib/aarch64-linux-gnu/libnvparsers.so /usr/lib/aarch64-linux-gnu/libnvonnxparser.so /usr/lib/aarch64-linux-gnu/libnvparsers.so -Wl,--unresolved-symbols=ignore-in-shared-libs
CMakeFiles/trtexec.dir/trtexec.cpp.o: In function `main':
trtexec.cpp:(.text+0xecc): undefined reference to `initLibNvInferPlugins'
collect2: error: ld returned 1 exit status
samples/trtexec/CMakeFiles/trtexec.dir/build.make:184: recipe for target 'out/trtexec' failed
make[2]: *** [out/trtexec] Error 1
make[2]: Leaving directory '/home/agrosy/git/TRT_docker/TensorRT/build'
CMakeFiles/Makefile2:1789: recipe for target 'samples/trtexec/CMakeFiles/trtexec.dir/all' failed
make[1]: *** [samples/trtexec/CMakeFiles/trtexec.dir/all] Error 2
make[1]: Leaving directory '/home/agrosy/git/TRT_docker/TensorRT/build'
Makefile:135: recipe for target 'all' failed
make: *** [all] Error 2
- There is a similar issue https://github.com/NVIDIA/TensorRT/issues/1308#issue-923372506 but it was not a useful solution.
Could someone help me shed some light on this please.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with samples/trtexec/trtexec.cpp and the generated samples/trtexec/CMakeFiles/trtexec.dir/link.txt shown in the report. Reproduce the release/8.0 native aarch64 build with the provided CMake and make commands, compare the linking setup with issue 1308, and consider the issue done when trtexec links successfully and the full build completes.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cmake, cpp
- Domain
- build-system
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 38/100