intel / torch-ccl

oneCCL Bindings for Pytorch*
BSD 3-Clause "New" or "Revised" License
86 stars 25 forks source link

doesn't work on CPU only environment #52

Open manjeetbhati opened 1 year ago

manjeetbhati commented 1 year ago

The documentation is misleading https://github.com/intel/torch-ccl#install-from-source

here it says for cpu only backend python setup.py install

when its built on cpu only environment it looks for cuda related libraries and fails

below is the trace

-- Detecting CXX compile features - done CUDA_TOOLKIT_ROOT_DIR not found or specified -- Could NOT find CUDA (missing: CUDA_TOOLKIT_ROOT_DIR CUDA_NVCC_EXECUTABLE CUDA_INCLUDE_DIRS CUDA_CUDART_LIBRARY) CMake Warning at /usr/local/lib/python3.10/dist-packages/torch/share/cmake/Caffe2/public/cuda.cmake:31 (message): Caffe2: CUDA cannot be found. Depending on whether you are building Caffe2 or a Caffe2 dependent library, the next warning / error will give you more info. Call Stack (most recent call first): /usr/local/lib/python3.10/dist-packages/torch/share/cmake/Caffe2/Caffe2Config.cmake:88 (include) /usr/local/lib/python3.10/dist-packages/torch/share/cmake/Torch/TorchConfig.cmake:68 (find_package) CMakeLists.txt:29 (find_package)

CMake Error at /usr/local/lib/python3.10/dist-packages/torch/share/cmake/Caffe2/Caffe2Config.cmake:90 (message): Your installed Caffe2 version uses CUDA but I cannot find the CUDA libraries. Please set the proper CUDA prefixes and / or install CUDA. Call Stack (most recent call first): /usr/local/lib/python3.10/dist-packages/torch/share/cmake/Torch/TorchConfig.cmake:68 (find_package) CMakeLists.txt:29 (find_package)

-- Configuring incomplete, errors occurred! Traceback (most recent call last): File "/root/openai/torch-ccl/setup.py", line 213, in setup( File "/usr/lib/python3/dist-packages/setuptools/init.py", line 153, in setup return distutils.core.setup(**attrs) File "/usr/lib/python3.10/distutils/core.py", line 148, in setu

NaamaKadosh commented 1 year ago

Hi, I encountered exactly the same issue. Is there a workaround? Thanks