Description
Cmake step of tensorRT compile fails because zlib cannot be found.
Environment
TensorRT Version: -
Hardware: Jetson Xavier
Jetpack: 4.4
CUDA Version: 10.2
CUDNN Version: 8.0
Baremetal or Container (if container which image + tag): Using custom container using nvidia l4t repo
Relevant Files
CMakeError.log: CMakeError.log (3.0 KB)
Console output:
/workspace/cmake-3.24.0-rc4-linux-aarch64/bin/cmake .. -DTRT_LIB_DIR=$TRT_RELEASE/lib -DTRT_OUT_DIR=`pwd`/out -DTRT_INC_DIR=$TRT_RELEASE/include -DTRT_PLATFORM_ID=aarch64 -DCUDA_VERSION=10.2 -DBUILD_SAMPLES=OFF -DZLIB_LIBRARY=/u
sr/lib/x86_64-linux-gnu/libz.so
Building for TensorRT version: 7.2.1, library version: 7
-- The CXX compiler identification is GNU 7.5.0
-- The CUDA compiler identification is NVIDIA 10.2.89
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/g++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Detecting CUDA compiler ABI info
-- Detecting CUDA compiler ABI info - done
-- Check for working CUDA compiler: /usr/local/cuda-10.2/bin/nvcc - skipped
-- Detecting CUDA compile features
-- Detecting CUDA compile features - done
-- Targeting TRT Platform: aarch64
-- CUDA version set to 10.2
-- cuDNN version set to 8.0
-- Protobuf version set to 3.0.0
-- Checking for one of the modules 'zlib'
CMake Error at /workspace/cmake-3.24.0-rc4-linux-aarch64/share/cmake-3.24/Modules/FindPkgConfig.cmake:894 (message):
None of the required 'zlib' found
Call Stack (most recent call first):
third_party/zlib.cmake:18 (pkg_search_module)
CMakeLists.txt:98 (include)
I have installed zlib using apt install zlib1g-dev and also pkg-config. Those install steps are not in the dockerfile(s) shared below.
Steps To Reproduce
git clone -b master https://github.com/nvidia/TensorRT TensorRT
cd TensorRT
git submodule update --init --recursive
export TRT_SOURCE=`pwd`
export TRT_RELEASE=`pwd`/TensorRT-7.2.1.6
cd $TRT_SOURCE
mkdir -p build && cd build
cmake .. -DTRT_LIB_DIR=$TRT_RELEASE/lib -DTRT_OUT_DIR=`pwd`/out -DTRT_PLATFORM_ID=aarch64 -DCUDA_VERSION=10.2
make -j$(nproc)
The above steps are done in a container.
base dockerfile: run make image after cloning nvidia l4t repo
TRT build docker: Dockerfile_ubuntu20_l4t_infer_trt_7_jp44 (3.3 KB)