Some error in setup BpOps，“BP-Net-main/exts/build/temp.linux-x86_64-cpython-39/bp_cuda_kernel.o: No such file or directory”

[yudmoe]

I installed cuda12.1 on the server in my lab, and the gcc version is 12.3. After creating an environment named bp according to the README, I first encountered the ninja problem "Command '['ninja', '-v']' returned non-zero exit status 1". After changing command = ['ninja', '-v'] to command = ['ninja', '--version'] according to this issue.
I encountered the following error(zzy is my acount name):

(bp) zzy@ICML:~/code/BP-Net-main/exts$ python setup.py install
running install
running bdist_egg
running egg_info
writing BpOps.egg-info/PKG-INFO
writing dependency_links to BpOps.egg-info/dependency_links.txt
writing top-level names to BpOps.egg-info/top_level.txt
reading manifest file 'BpOps.egg-info/SOURCES.txt'
writing manifest file 'BpOps.egg-info/SOURCES.txt'
installing library code to build/bdist.linux-x86_64/egg
running install_lib
running build_ext
/home/zzy/anaconda3/envs/bp/lib/python3.9/site-packages/torch/utils/cpp_extension.py:426: UserWarning: There are no /home/zzy/anaconda3/envs/bp/bin/x86_64-conda-linux-gnu-c++ version bounds defined for CUDA version 12.1
warnings.warn(f'There are no {compiler_name} version bounds defined for CUDA version {cuda_str_version}')
building 'BpOps' extension
Emitting ninja build file /home/zzy/code/BP-Net-main/exts/build/temp.linux-x86_64-cpython-39/build.ninja...
Compiling objects...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
1.10.1
creating build/lib.linux-x86_64-cpython-39
/home/zzy/anaconda3/envs/bp/bin/x86_64-conda-linux-gnu-c++ -shared -Wl,--allow-shlib-undefined -Wl,-rpath,/home/zzy/anaconda3/envs/bp/lib -Wl,-rpath-link,/home/zzy/anaconda3/envs/bp/lib -L/home/zzy/anaconda3/envs/bp/lib -Wl,--allow-shlib-undefined -Wl,-rpath,/home/zzy/anaconda3/envs/bp/lib -Wl,-rpath-link,/home/zzy/anaconda3/envs/bp/lib -L/home/zzy/anaconda3/envs/bp/lib -Wl,-O2 -Wl,--sort-common -Wl,--as-needed -Wl,-z,relro -Wl,-z,now -Wl,--disable-new-dtags -Wl,--gc-sections -Wl,--allow-shlib-undefined -Wl,-rpath,/home/zzy/anaconda3/envs/bp/lib -Wl,-rpath-link,/home/zzy/anaconda3/envs/bp/lib -L/home/zzy/anaconda3/envs/bp/lib -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -ffunction-sections -pipe -isystem /home/zzy/anaconda3/envs/bp/include -DNDEBUG -D_FORTIFY_SOURCE=2 -O2 -isystem /home/zzy/anaconda3/envs/bp/include /home/zzy/code/BP-Net-main/exts/build/temp.linux-x86_64-cpython-39/bp_cuda.o /home/zzy/code/BP-Net-main/exts/build/temp.linux-x86_64-cpython-39/bp_cuda_kernel.o -L/home/zzy/anaconda3/envs/bp/lib/python3.9/site-packages/torch/lib -L/usr/local/cuda-12.1/lib64 -lc10 -ltorch -ltorch_cpu -ltorch_python -lcudart -lc10_cuda -ltorch_cuda -o build/lib.linux-x86_64-cpython-39/BpOps.cpython-39-x86_64-linux-gnu.so
/home/zzy/anaconda3/envs/bp/bin/../lib/gcc/x86_64-conda-linux-gnu/12.3.0/../../../../x86_64-conda-linux-gnu/bin/ld: cannot find /home/zzy/code/BP-Net-main/exts/build/temp.linux-x86_64-cpython-39/bp_cuda_kernel.o: No such file or directory
collect2: error: ld returned 1 exit status
error: command '/home/zzy/anaconda3/envs/bp/bin/x86_64-conda-linux-gnu-c++' failed with exit code 1

This seems to be because the bp_cuda_kernel.o file is not generated during compilation. Is it because of my cuda version?

[kakaxi314]

Sorry , I do not know this issue either.