Rocky Linux 8.5 安装带CUDA版本的Opencv4.5.5

最新推荐文章于 2025-06-04 11:43:23 发布

蓝风23

最新推荐文章于 2025-06-04 11:43:23 发布

阅读量1.1k

点赞数

CC 4.0 BY-SA版权

分类专栏：机器视觉高性能服务调试技巧文章标签： linux opencv 运维

本文链接：https://blog.csdn.net/bluewind_1988/article/details/124270367

调试技巧同时被 3 个专栏收录

15 篇文章

订阅专栏

高性能服务

13 篇文章

订阅专栏

机器视觉

12 篇文章

订阅专栏

本文详细介绍了CUDA 11.2的测试过程，包括两个GPU的性能指标，以及如何源码安装ffmpeg和opencv 4.5.5，重点展示了opencv的CUDA、TBB、FFmpeg和Python支持的配置步骤。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

一、CUDA测试

//找到NVIDIA_CUDA-11.2_Samples
cd NVIDIA_CUDA-11.2_Samples/
cd 1_Utilities/
cd deviceQuery
make
./deviceQuery
./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 2 CUDA Capable device(s)

Device 0: "GeForce GTX 1060 6GB"
  CUDA Driver Version / Runtime Version          11.2 / 11.2
  CUDA Capability Major/Minor version number:    6.1
  Total amount of global memory:                 6078 MBytes (6373572608 bytes)
  (10) Multiprocessors, (128) CUDA Cores/MP:     1280 CUDA Cores
  GPU Max Clock rate:                            1759 MHz (1.76 GHz)
  Memory Clock rate:                             4004 Mhz
  Memory Bus Width:                              192-bit
  L2 Cache Size:                                 1572864 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
  Maximum Layered 1D Texture Size, (num) layers  1D=(32768), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(32768, 32768), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total shared memory per multiprocessor:        98304 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 2 copy engine(s)
  Run time limit on kernels:                     No
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device supports Managed Memory:                Yes
  Device supports Compute Preemption:            Yes
  Supports Cooperative Kernel Launch:            Yes
  Supports MultiDevice Co-op Kernel Launch:      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 3 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

Device 1: "GeForce GTX 1060 6GB"
  CUDA Driver Version / Runtime Version          11.2 / 11.2
  CUDA Capability Major/Minor version number:    6.1
  Total amount of global memory:                 6078 MBytes (6373572608 bytes)
  (10) Multiprocessors, (128) CUDA Cores/MP:     1280 CUDA Cores
  GPU Max Clock rate:                            1759 MHz (1.76 GHz)
  Memory Clock rate:                             4004 Mhz
  Memory Bus Width:                              192-bit
  L2 Cache Size:                                 1572864 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
  Maximum Layered 1D Texture Size, (num) layers  1D=(32768), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(32768, 32768), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total shared memory per multiprocessor:        98304 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 2 copy engine(s)
  Run time limit on kernels:                     No
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device supports Managed Memory:                Yes
  Device supports Compute Preemption:            Yes
  Supports Cooperative Kernel Launch:            Yes
  Supports MultiDevice Co-op Kernel Launch:      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 130 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
> Peer access from GeForce GTX 1060 6GB (GPU0) -> GeForce GTX 1060 6GB (GPU1) : No
> Peer access from GeForce GTX 1060 6GB (GPU1) -> GeForce GTX 1060 6GB (GPU0) : No

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 11.2, CUDA Runtime Version = 11.2, NumDevs = 2
Result = PASS

二、opencv安装

1、源码安装ffmpeg

yum install cmake gcc gcc-c++

wget http://www.tortall.net/projects/yasm/releases/yasm-1.3.0.tar.gz
tar xf yasm-1.3.0.tar.gz
cd yasm-1.3.0/
./configure --prefix=/usr/local
make -j8 
make install 

wget https://johnvansickle.com/ffmpeg/release-source/ffmpeg-4.1.tar.xz
tar xf ffmpeg-4.1.tar.xz
cd ffmpeg-4.1/
./configure --enable-shared  --prefix=/opt/ffmpeg
make -j8
make install
echo "/opt/ffmpeg/lib" >> /etc/ld.so.conf
ldconfig
（
ldconfig: /usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_ops_train.so.8 不是符号链接

ldconfig: /usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_ops_infer.so.8 不是符号链接

ldconfig: /usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_cnn_train.so.8 不是符号链接

ldconfig: /usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn.so.8 不是符号链接

ldconfig: /usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_cnn_infer.so.8 不是符号链接

ldconfig: /usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_adv_train.so.8 不是符号链接

ldconfig: /usr/local/cuda-11.2/targets/x86_64-linux/lib/libcudnn_adv_infer.so.8 不是符号链接
）
//均如下解决：
mv libcudnn_ops_infer.so libcudnn_ops_infer.so.bk
mv libcudnn_ops_infer.so.8 libcudnn_ops_infer.so.8.bk
ln -sf libcudnn_ops_infer.so.8.1.1 libcudnn_ops_infer.so.8
ln -sf libcudnn_ops_infer.so.8 libcudnn_ops_infer.so

配置环境变量

vi /etc/profile
//末尾添加（已修正）：
export PATH=/opt/ffmpeg/bin:$PATH
//使配置生效
source /etc/profile
(设置环境变量出了问题，导致su ls等常用命令不能用：
 由于无法使用常用命令，故直接在shell下不能够修改，这时就需要设置临时环境变量PATH。export PATH="/usr/sbin:/usr/bin:/usr/local/bin:/usr/local/sbin:/bin:/sbin"
    然后就可以正常使用命令了，这时就要修改导致PATH出问题的文件。由于我是修改的/etc/profile配置文件，所以要进去重新修改。修改之后为了使其生效，需要输入source /etc/profile,或者重启。）

2、安装一些依赖包

//替换阿里Yum源
sed -e 's|^mirrorlist=|#mirrorlist=|g' -e 's|^#baseurl=http://dl.rockylinux.org/$contentdir|baseurl=https://mirrors.aliyun.com/rockylinux|g' -i.bak /etc/yum.repos.d/Rocky-*.repo
//makecache一下
dnf makecache

yum install -y cmake gcc gtk2-devel unzip 
yum install -y libwebp-devel libjpeg-turbo-devel
yum install -y freeglut-devel mesa-libGL mesa-libGL-devel libtiff-devel
yum install -y tbb-devel
yum install -y boost boost-thread boost-devel
yum install -y gtk+-devel gimp-devel gimp-devel-tools gimp-help-browser zlib-devel libtiff-devel libjpeg-devel libpng-devel gstreamer-devel libraw1394-devel libdc1394-devel jasper-devel jasper-utils swig python libtool nasm libv4l-devel libv4l-devel  python-devel numpy gstreamer-plugins-base-devel gtk2-devel gtk2-devel-docs  libavcodec-devel libavformat-devel libswscale-devel libavutil-devel  libeigen3-devel libtbb-devel libtiff-dev libavformat-devel libpq-devel  libxine2-devel libglew-devel libtiff5-devel gstreamer-plugins-base-devel libjpeg-turbo-devel jasper-devel openexr-devel tbb-devel --skip-broken

3、安装opencv

wget https://github.com/opencv/opencv/archive/4.5.5.zip
unzip opencv-4.5.5.zip
mkdir opencv_build

安装包放置目录如下：
/opencv
    /opencv-4.5.1           # opencv 源码目录
    /opencv_contrib-4.5.1   # opencv_contrib 源码目录
    /opencv_build           # opencv 编译结果目录

cd opencv_build
cmake 参数如下。
make -j30
make install

cmake的内容:

CUDA_ARCH_BIN: gpu 算力，参考：[https://developer.nvidia.com/zh-cn/cuda-gpus]

cmake \
    -D CMAKE_BUILD_TYPE=RELEASE \
    -D CMAKE_C_COMPILER=/usr/bin/gcc \
    -D CMAKE_INSTALL_PREFIX=$(python3 -c "import sys; print(sys.prefix)") \
    -D OPENCV_EXTRA_MODULES_PATH=../opencv_contrib-4.5.5/modules \
    -D CUDA_CUDA_LIBRARY=/usr/local/cuda-11.2/targets/x86_64-linux/lib/stubs/libcuda.so \
    -D CUDA_ARCH_BIN=6.1 \
    -D CUDA_ARCH_PTX="" \
    -D WITH_CUDA=ON \
    -D WITH_TBB=ON \
    -D WITH_FFMPEG=ON \
    -D BUILD_PYTHON_SUPPORT=ON \
    -D BUILD_NEW_PYTHON_SUPPORT=ON \
    -D BUILD_OPENCV_PYTHON3=ON \
    -D PYTHON3_INCLUDE_DIR=$(python3 -c "from distutils.sysconfig import get_python_inc; print(get_python_inc())") \
    -D PYTHON3_PACKAGES_PATH=$(python3 -c "from distutils.sysconfig import get_python_lib; print(get_python_lib())") \
    -D WITH_V4L=ON \
    -D INSTALL_C_EXAMPLES=ON \
    -D INSTALL_PYTHON_EXAMPLES=ON \
    -D PYTHON3_EXECUTABLE=$(which python3) \
    -D BUILD_EXAMPLES=ON \
    -D WITH_QT=ON \
    -D WITH_GSTREAMER=ON \
    -D WITH_OPENGL=ON \
    -D ENABLE_FAST_MATH=1 \
    -D CUDA_FAST_MATH=1 \
    -D OPENCV_GENERATE_PKGCONFIG=ON \
    -D OPENCV_PC_FILE_NAME=opencv.pc \
    -D CUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda \
    -D CMAKE_LIBRARY_PATH=/usr/local/cuda-11.2/targets/x86_64-linux/lib/stubs \
    -D WITH_CUBLAS=ON \
    -D WITH_NVCUVID=ON \
    -D BUILD_opencv_cudacodec=ON \
    -D OPENCV_DNN_CUDA=ON \
    -D WITH_CUDNN=ON \
    -D OPENCV_ENABLE_NONFREE=ON\
    -D WITH_GSTREAMER=ON \
    -D BUILD_EXAMPLES=ON  \
    ../opencv-4.5.5