这里使用cmake配置编译文件，ninja进行编译，建议自己上网了解cmake，比如看菜鸟教程
*sudo apt install* build-essential vim wget curl cmake-qt-gui ninja-build python3-dev python3 python3-venv python3-pip python3-numpy python3-scipy python3-matplotlib python3-pandas python3-full jupyter jupyter-nbconvert jupyter-notebook python3-ipykernel jupyter-qtconsole libx11-dev libxpm-dev libxft-dev libxext-dev libpng-dev libjpeg-dev libssl-dev uuid-dev libpcre3-dev libglu1-mesa-dev libglew-dev libftgl-dev default-libmysqlclient-dev libfftw3-dev libcfitsio-dev libavahi-compat-libdnssd-dev libldap-dev libxml2-dev libkrb5-dev libopenblas-dev libtbb-dev libgsl-dev libunuran-dev liburing-dev libgraphviz-dev libgif-dev libafterimage-dev davix-dev libgl2ps-dev liblz4-dev lzma-dev qtwebengine5-dev nlohmann-json3-dev libvdt-dev xrootd-client xrootd-server libxrootd-client-dev libxrootd-server-dev libxxhash-dev git libxerces-c-dev libexpat1-dev qt3d5-dev libxmu-dev veccore-dev libglib2.0-dev gnulib libgmp-dev libmpc-dev libmpfr-dev libzstd-dev libeigen3-dev mpi-default-dev binutils-dev libdw-dev libdwarf-dev graphviz
复制粘贴到终端即可，这样root在编译时大部分的依赖已经安装好，不需要root安装（也防止root帮你安装依赖时配置的环境不干净）。
如果需要编译支持CUDA的ROOT，再安装以下这些包（假定你已经安装了nvidia-driver）：
sudo apt install nvidia-cuda-toolkit-gcc nvidia-cudnn
有些依赖在不同的Linux发行版中有略微区别，比如在较新的 Debian 版本中，libpcre3-dev可能已被libpcre2-dev取代。
上官网找源码下载，比如root_v6.32.06.source.tar.gz
Download the realeases for the desired platform and ROOT version
下载之后压缩包copy到放源码的文件夹（自己建），比如说~/app/，然后进到app文件夹解压
tar -xzvf root_v<version>.source.tar.gz
创建build文件夹
mkdir build
进入build文件夹，CMake生成构建文件