Failed to build oneflow inside your docker

[cam-ai]

Summary

I downloaded your project and docker, when i try to build oneflow inside your docker, it failed.

Code to reproduce bug

cmake .. -C ../cmake/caches/cn/cuda.cmake -DCMAKE_CUDA_ARCHITECTURES=80 -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda -DCUDNN_ROOT_DIR=/usr/lib64

CMake Error at /opt/_internal/pipx/venvs/cmake/lib/python3.10/site-packages/cmake/data/share/cmake-3.27/Modules/FetchContent.cmake:1662 (message): Build step for tbb failed: 2 Call Stack (most recent call first): /opt/_internal/pipx/venvs/cmake/lib/python3.10/site-packages/cmake/data/share/cmake-3.27/Modules/FetchContent.cmake:1802:EVAL:2 (__FetchContent_directPopulate) /opt/_internal/pipx/venvs/cmake/lib/python3.10/site-packages/cmake/data/share/cmake-3.27/Modules/FetchContent.cmake:1802 (cmake_language) /opt/_internal/pipx/venvs/cmake/lib/python3.10/site-packages/cmake/data/share/cmake-3.27/Modules/FetchContent.cmake:2016 (FetchContent_Populate) external/onetbb/CMakeLists.txt:14 (FetchContent_MakeAvailable)

System Information

ubuntu 20.04

• What is your OneFlow installation (pip, source, dockerhub):
• OS:
• OneFlow version (run python3 -m oneflow --doctor):
• Python version:
• CUDA driver version:
• GPU models:
• Other info:

[westfly]

@cam-ai Which Docker image do you employ?I test oneflowinc/oneflow:0.9.1.dev20240202-cuda12.1 works fine

docker run -it --rm --runtime=nvidia --privileged --network host oneflowinc/oneflow:0.9.1.dev20240202-cuda12.1  python3 -m oneflow --doctor

It appears that this image is a running image which has no gcc and nvcc rather than a build-dev image