lingyuzeng/cdc_dockerfile
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update

Browse Source
This commit is contained in:
lingyuzeng
2024-07-18 13:07:12 +08:00
parent bfe427194e
commit b7584b2a04
1 changed files with 4 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
finetune/Dockerfile.ngc
View File

@@ -20,23 +20,16 @@ ENV MLNX_OFED_VERSION=23.10-3.2.2.0
ENV STAGE_DIR=/tmp
RUN <<EOT
#!/bin/bash
# SYSTEM_NAME=$(lsb_release -cs) # 查看发行版本
# Pre-build **latest** DeepSpeed, so it would be ready for testing (otherwise, the 1st deepspeed test will timeout)
# This has to be run (again) inside the GPU VMs running the tests.
# The installation works here, but some tests fail, if we do not pre-build deepspeed again in the VMs running the tests.
# TODO: Find out why test fail. install deepspeed
# DS_BUILD_CPU_ADAM=${DS_BUILD_CPU_ADAM} DS_BUILD_FUSED_ADAM={DS_BUILD_FUSED_ADAM} python3 -m pip install "deepspeed<=0.14.0" --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check 2>&1
# from https://github.com/huggingface/transformers/blob/main/docker/transformers-pytorch-deepspeed-latest-gpu/Dockerfile install deepspeed fail
# reference deepspeed install from https://github.com/microsoft/DeepSpeed/blob/master/docker/Dockerfile
# install Mellanox OFED prepare
apt-get update
apt install libnvidia-compute-535
apt install -y libnvidia-compute-535
apt-get install -y pciutils tk kmod libusb-1.0-0 tcl chrpath libpci3 bison lsof graphviz ethtool swig udev libltdl-dev libelf1 libmnl0 debhelper flex libfuse2
# install Mellanox OFED
mkdir -p ${STAGE_DIR}
wget -q -O - http://www.mellanox.com/downloads/ofed/MLNX_OFED-${MLNX_OFED_VERSION}/MLNX_OFED_LINUX-${MLNX_OFED_VERSION}-ubuntu22.04-x86_64.tgz | tar xzf - -C ${STAGE_DIR}
wget http://www.mellanox.com/downloads/ofed/MLNX_OFED-${MLNX_OFED_VERSION}/MLNX_OFED_LINUX-${MLNX_OFED_VERSION}-ubuntu22.04-x86_64.tgz -O ${STAGE_DIR}/MLNX_OFED_LINUX-${MLNX_OFED_VERSION}-ubuntu22.04-x86_64.tgz
tar xzf ${STAGE_DIR}/MLNX_OFED_LINUX-${MLNX_OFED_VERSION}-ubuntu22.04-x86_64.tgz -C ${STAGE_DIR}
cd ${STAGE_DIR}/MLNX_OFED_LINUX-${MLNX_OFED_VERSION}-ubuntu22.04-x86_64
./mlnxofedinstall --user-space-only --without-fw-update --all -q
./mlnxofedinstall --user-space-only --without-fw-update --all -q > ${STAGE_DIR}/mlnxofedinstall.log 2>&1
cd ${STAGE_DIR}
rm -rf ${STAGE_DIR}/MLNX_OFED_LINUX-${MLNX_OFED_VERSION}-ubuntu22.04-x86_64*
EOT