[2026春季][T1-1-2] acoier

[lihaokun-2026]

环境初始化

英伟达平台

cat >> ~/.bashrc <<'EOF'
# ===== NVIDIA CUDA ENV =====
export CUDA_HOME=/usr/local/cuda
export CUDA_PATH=$CUDA_HOME
export PATH=$CUDA_HOME/bin:$PATH
export LD_LIBRARY_PATH=$CUDA_HOME/lib64:$LD_LIBRARY_PATH
# export CUDA_VISIBLE_DEVICES=0
EOF
source ~/.bashrc

沐曦平台

#!/usr/bin/env bash
export MACA_PATH=/opt/maca
export MACA_CLANG_PATH=${MACA_PATH}/mxgpu_llvm/bin
export PATH=${MACA_PATH}/mxgpu_llvm/bin:${MACA_PATH}/bin:${PATH}
export LD_LIBRARY_PATH=${MACA_PATH}/lib:${MACA_PATH}/mxgpu_llvm/lib:${MACA_PATH}/ompi/lib:${LD_LIBRARY_PATH}
# PyTorch Extension / cu-bridge
export CUDA_PATH=${MACA_PATH}/tools/cu-bridge/
export CUCC_PATH=${MACA_PATH}/tools/cu-bridge/
export PATH=${CUCC_PATH}/tools:${CUCC_PATH}/bin:${PATH}

天数平台

#!/usr/bin/env bash
export COREX_HOME=/usr/local/corex
export PATH=$COREX_HOME/bin:$PATH
export LD_LIBRARY_PATH=$COREX_HOME/lib:$COREX_HOME/lib64:$LD_LIBRARY_PATH

第三方库安装

pip install ninetoothed
pip install -v -e . #对ntops项目进行编译安装

不同平台测试结果

nv平台

沐曦平台

天数平台