Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

An error occurred when installing torch-quiver. #121

Closed
l1nkr opened this issue Feb 25, 2022 · 3 comments
Closed

An error occurred when installing torch-quiver. #121

l1nkr opened this issue Feb 25, 2022 · 3 comments

Comments

@l1nkr
Copy link
Contributor

l1nkr commented Feb 25, 2022

I tried to install torch-quiver, but i got an error as below.

[My environment]

OS : Linux version 3.10.0-1127.el7.x86_64 (mockbuild@kbuilder.bsys.centos.org) (gcc version 4.8.5 20150623 (Red Hat 4.8.5-39) (GCC) ) #1 SMP Tue Mar 31 23:36:51 UTC 2020

gpu : 2080ti

cuda : release 10.2, V10.2.89

python : 3.9.7 in conda

pytorch : '1.10.2' in conda

pyg : 2.0.3 in conda

gcc : 6.5.0

run pip install torch-quiver

Collecting torch-quiver
  Using cached torch_quiver-0.1.1.tar.gz (117 kB)
Building wheels for collected packages: torch-quiver
  Building wheel for torch-quiver (setup.py) ... error
  ERROR: Command errored out with exit status 1:
   command: /home/xxx/anaconda3/envs/torch_quiver/bin/python -u -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-kokivq60/torch-quiver_c32f9749bbf34bd6b524a81b30256cd0/setup.py'"'"'; __file__='"'"'/tmp/pip-install-kokivq60/torch-quiver_c32f9749bbf34bd6b524a81b30256cd0/setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(__file__) if os.path.exists(__file__) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' bdist_wheel -d /tmp/pip-wheel-h6bay53m
       cwd: /tmp/pip-install-kokivq60/torch-quiver_c32f9749bbf34bd6b524a81b30256cd0/
  Complete output (40 lines):
  /home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/distutils/extension.py:131: UserWarning: Unknown Extension options: 'with_cuda'
    warnings.warn(msg)
  running bdist_wheel
  /home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/utils/cpp_extension.py:381: UserWarning: Attempted to use ninja as the BuildExtension backend but we could not find ninja.. Falling back to using the slow distutils backend.
    warnings.warn(msg.format('we could not find ninja.'))
  running build
  running build_py
  creating build
  creating build/lib.linux-x86_64-3.9
  creating build/lib.linux-x86_64-3.9/quiver
  copying ./srcs/python/quiver/partition.py -> build/lib.linux-x86_64-3.9/quiver
  copying ./srcs/python/quiver/comm.py -> build/lib.linux-x86_64-3.9/quiver
  copying ./srcs/python/quiver/feature.py -> build/lib.linux-x86_64-3.9/quiver
  copying ./srcs/python/quiver/shard_tensor.py -> build/lib.linux-x86_64-3.9/quiver
  copying ./srcs/python/quiver/async_cuda_sampler.py -> build/lib.linux-x86_64-3.9/quiver
  copying ./srcs/python/quiver/__init__.py -> build/lib.linux-x86_64-3.9/quiver
  copying ./srcs/python/quiver/utils.py -> build/lib.linux-x86_64-3.9/quiver
  creating build/lib.linux-x86_64-3.9/quiver/multiprocessing
  copying ./srcs/python/quiver/multiprocessing/reductions.py -> build/lib.linux-x86_64-3.9/quiver/multiprocessing
  copying ./srcs/python/quiver/multiprocessing/__init__.py -> build/lib.linux-x86_64-3.9/quiver/multiprocessing
  creating build/lib.linux-x86_64-3.9/quiver/pyg
  copying ./srcs/python/quiver/pyg/__init__.py -> build/lib.linux-x86_64-3.9/quiver/pyg
  copying ./srcs/python/quiver/pyg/sage_sampler.py -> build/lib.linux-x86_64-3.9/quiver/pyg
  running build_ext
  building 'torch_quiver' extension
  creating build/temp.linux-x86_64-3.9
  creating build/temp.linux-x86_64-3.9/srcs
  creating build/temp.linux-x86_64-3.9/srcs/cpp
  creating build/temp.linux-x86_64-3.9/srcs/cpp/src
  creating build/temp.linux-x86_64-3.9/srcs/cpp/src/quiver
  creating build/temp.linux-x86_64-3.9/srcs/cpp/src/quiver/cpu
  creating build/temp.linux-x86_64-3.9/srcs/cpp/src/quiver/cuda
  creating build/temp.linux-x86_64-3.9/srcs/cpp/src/quiver/torch
  gcc -pthread -B /home/xxx/anaconda3/envs/torch_quiver/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /home/xxx/anaconda3/envs/torch_quiver/include -I/home/xxx/anaconda3/envs/torch_quiver/include -fPIC -O2 -isystem /home/xxx/anaconda3/envs/torch_quiver/include -fPIC -I/tmp/pip-install-kokivq60/torch-quiver_c32f9749bbf34bd6b524a81b30256cd0/./srcs/cpp/include -I/usr/local/cuda/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/torch/csrc/api/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/TH -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/THC -I/home/xxx/anaconda3/envs/torch_quiver/include/python3.9 -c srcs/cpp/src/quiver/cpu/tensor.cpp -o build/temp.linux-x86_64-3.9/srcs/cpp/src/quiver/cpu/tensor.o -std=c++17 -DHAVE_CUDA -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=torch_quiver -D_GLIBCXX_USE_CXX11_ABI=0
  /home/xxx/cuda10.2/bin/nvcc -I/tmp/pip-install-kokivq60/torch-quiver_c32f9749bbf34bd6b524a81b30256cd0/./srcs/cpp/include -I/usr/local/cuda/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/torch/csrc/api/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/TH -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/THC -I/home/xxx/anaconda3/envs/torch_quiver/include/python3.9 -c srcs/cpp/src/quiver/cuda/quiver_comm.cu -o build/temp.linux-x86_64-3.9/srcs/cpp/src/quiver/cuda/quiver_comm.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options '-fPIC' -O3 --expt-extended-lambda -lnuma -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=torch_quiver -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_75,code=compute_75 -gencode=arch=compute_75,code=sm_75 -std=c++14
  /home/xxx/cuda10.2/bin/nvcc -I/tmp/pip-install-kokivq60/torch-quiver_c32f9749bbf34bd6b524a81b30256cd0/./srcs/cpp/include -I/usr/local/cuda/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/torch/csrc/api/include -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/TH -I/home/xxx/anaconda3/envs/torch_quiver/lib/python3.9/site-packages/torch/include/THC -I/home/xxx/anaconda3/envs/torch_quiver/include/python3.9 -c srcs/cpp/src/quiver/cuda/quiver_feature.cu -o build/temp.linux-x86_64-3.9/srcs/cpp/src/quiver/cuda/quiver_feature.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options '-fPIC' -O3 --expt-extended-lambda -lnuma -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="_gcc" -DPYBIND11_STDLIB="_libstdcpp" -DPYBIND11_BUILD_ABI="_cxxabi1011" -DTORCH_EXTENSION_NAME=torch_quiver -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_75,code=compute_75 -gencode=arch=compute_75,code=sm_75 -std=c++14
  srcs/cpp/src/quiver/cuda/quiver_feature.cu(192): warning: arithmetic on pointer to void or function type

  nvcc error   : 'cicc' died due to signal 11 (Invalid memory reference)
  error: command '/home/xxx/cuda10.2/bin/nvcc' failed with exit code 11
  ----------------------------------------
  ERROR: Failed building wheel for torch-quiver

Thanks.

@l1nkr
Copy link
Contributor Author

l1nkr commented Feb 26, 2022

-I/usr/local/cuda/include

Maybe that's why it failed. I don't have root privileges so I just update cuda in user level . How to solve this problem

@ZenoTan
Copy link
Member

ZenoTan commented Feb 27, 2022

I think the problem is about the compiler. Could you try nvcc to compile other programs? You can set your cuda directory here when you build from source.

@l1nkr
Copy link
Contributor Author

l1nkr commented Feb 28, 2022

Thanks. I have solved this problem. It also needs to be modified here.

@l1nkr l1nkr closed this as completed Feb 28, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants