昇腾NPU环境搭建
如果进入服务器输入npu-smi info
可以看到npu情况,请直接跳转第三步
STEP1: 服务器安装依赖
sudo yum install -y gcc gcc-c++ make cmake unzip zlib-devel libffi-devel openssl-devel pciutils net-tools sqlite-devel lapack-devel gcc-gfortran python3-develyum install dkmsyum install docker
STEP2: 下载驱动
https://www.hiascend.com/hardware/firmware-drivers/community?product=4&model=26&cann=8.2.RC1.alpha001&driver=Ascend+HDK+25.0.RC1
进入昇腾社区下载稳定驱动
下载两个驱动包,并上传服务器。
上传完毕后需要给两个文件root用户权限。
chmod 777 ./Ascend*
STEP 3: 创建用户
sudo groupadd HwHiAiUser
sudo useradd -g HwHiAiUser -d /home/HwHiAiUser -m HwHiAiUser -s /bin/bash
sudo usermod -aG HwHiAiUser $USER
STEP 4:安装驱动
sudo sh Ascend-hdk-910b-npu-driver_***.run --full --install-for-all
安装时若出现以下回显:
[WARNING]rebuild ko has something wrong, detail in /var/log/ascend_seclog/ascend_rebuild.log
Do you want to try build driver after input kernel absolute path? [y/n]:
输入y
并填入服务器的如下路径
/lib/modules/`uname -r`/build
结束后,输入npu-smi info
查看是否安装成功
STEP 5:安装firmware
sudo sh Ascend-hdk-910b-npu-firmware_***.run --full --install-for-all
STEP 6: 安装docker
https://www.hiascend.com/developer/ascendhub/detail/e26da9266559438b93354792f25b2f4a
用手机号登录即可
启动docker
docker run -it --shm-size=512g --name ‘mindspeed_llm’ --network host --privileged --device=/dev/davinci0 --device=/dev/davinci1 --device=/dev/davinci2 --device=/dev/davinci3 --device=/dev/davinci4 --device=/dev/davinci5 --device=/dev/davinci6 --device=/dev/davinci7 --device=/dev/davinci_manager --device=/dev/davinci_svm --device=/dev/hisi_hdc -v /usr/local/bin/npu-smi:/usr/local/bin/npu-smi -v /usr/local/sbin/npu-smi:/usr/local/sbin/npu-smi -v /usr/local/Ascend/driver:/usr/local/Ascend/driver -v /etc/ascend_install.info:/etc/ascend_install.info -v /usr/local/Ascend:/usr/local/Ascend -v /usr/local/Ascend/firmware:/usr/local/Ascend/firmware -v /usr/local/sbin/:/usr/local/sbin/ -v /home/:/home/ swr.cn-south-1.myhuaweicloud.com/ascendhub/mindspeed-llm:2024.rc4-arm /bin/bash