[Bug] pytorch2.0 isntall error: more than one user-defined conversion from "phalf" to "__half" applies:

[chenrui17]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• The bug has not been fixed in the latest version(https://github.com/open-mmlab/mmcv).

Environment

addict 2.4.0
certifi 2022.12.7
charset-normalizer 2.1.1
cmake 3.25.0
filelock 3.9.0
idna 3.4
mmcv-full 1.7.1
mpmath 1.2.1
networkx 3.0rc1
numpy 1.24.1
Pillow 9.3.0
pip 23.0
pytorch-triton 2.0.0+0d7e753227
requests 2.28.1
setuptools 67.1.0
sympy 1.11.1
torch 2.0.0.dev20230213+cu117
torchaudio 2.0.0.dev20230213+cu117
torchvision 0.15.0.dev20230213+cu117
typing_extensions 4.4.0
urllib3 1.26.13
wheel 0.38.4
yapf 0.32.0

Reproduces the problem - code sample

no

Reproduces the problem - command or script

shell script

# create conda env
conda create -n mmcv-test python=3.8 -y
# install pytorch2.0
pip3 install numpy --pre torch torchvision torchaudio --force-reinstall --index-url https://download.pytorch.org/whl/nightly/cu117
# clone github mmcv and install mmcv
MMCV_WITH_OPS=1 python setup.py install

Reproduces the problem - error message

error messages

[37/129] /usr/local/cuda/bin/nvcc  -DMMCV_WITH_CUDA -I/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/pytorch -I/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/common -I/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/common/cuda -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/TH -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/THC -I/usr/local/cuda/include -I/opt/conda/envs/torch-2.0-inductor-test/include/python3.8 -c -c /home/users/chenrui17/github/mmcv/mmcv/ops/csrc/pytorch/cuda/carafe_cuda.cu -o /home/users/chenrui17/github/mmcv/build/temp.linux-x86_64-cpython-38/mmcv/ops/csrc/pytorch/cuda/carafe_cuda.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options ''"'"'-fPIC'"'"'' -std=c++14 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=_ext -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_52,code=sm_52 -gencode=arch=compute_60,code=sm_60 -gencode=arch=compute_61,code=sm_61 -gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=sm_80 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86
FAILED: /home/users/chenrui17/github/mmcv/build/temp.linux-x86_64-cpython-38/mmcv/ops/csrc/pytorch/cuda/carafe_cuda.o
/usr/local/cuda/bin/nvcc  -DMMCV_WITH_CUDA -I/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/pytorch -I/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/common -I/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/common/cuda -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/TH -I/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/THC -I/usr/local/cuda/include -I/opt/conda/envs/torch-2.0-inductor-test/include/python3.8 -c -c /home/users/chenrui17/github/mmcv/mmcv/ops/csrc/pytorch/cuda/carafe_cuda.cu -o /home/users/chenrui17/github/mmcv/build/temp.linux-x86_64-cpython-38/mmcv/ops/csrc/pytorch/cuda/carafe_cuda.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options ''"'"'-fPIC'"'"'' -std=c++14 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=_ext -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_52,code=sm_52 -gencode=arch=compute_60,code=sm_60 -gencode=arch=compute_61,code=sm_61 -gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=sm_80 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86
/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/common/cuda/carafe_cuda_kernel.cuh(62): error: more than one user-defined conversion from "phalf" to "__half" applies:
            function "__half::__half(float)"
/usr/local/cuda/include/cuda_fp16.hpp(201): here
            function "__half::__half(__half &&)" (declared implicitly)

/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/c10/util/irange.h(54): warning #186-D: pointless comparison of unsigned integer with zero
          detected during:
            instantiation of "__nv_bool c10::detail::integer_iterator<I, one_sided, <unnamed>>::operator==(const c10::detail::integer_iterator<I, one_sided, <unnamed>> &) const [with I=size_t, one_sided=false, <unnamed>=0]"
(61): here
            instantiation of "__nv_bool c10::detail::integer_iterator<I, one_sided, <unnamed>>::operator!=(const c10::detail::integer_iterator<I, one_sided, <unnamed>> &) const [with I=size_t, one_sided=false, <unnamed>=0]"
/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/c10/core/TensorImpl.h(77): here

/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/c10/util/irange.h(54): warning #186-D: pointless comparison of unsigned integer with zero
          detected during:
            instantiation of "__nv_bool c10::detail::integer_iterator<I, one_sided, <unnamed>>::operator==(const c10::detail::integer_iterator<I, one_sided, <unnamed>> &) const [with I=std::size_t, one_sided=true, <unnamed>=0]"
(61): here
            instantiation of "__nv_bool c10::detail::integer_iterator<I, one_sided, <unnamed>>::operator!=(const c10::detail::integer_iterator<I, one_sided, <unnamed>> &) const [with I=std::size_t, one_sided=true, <unnamed>=0]"
/opt/conda/envs/torch-2.0-inductor-test/lib/python3.8/site-packages/torch/include/ATen/core/qualified_name.h(73): here

1 error detected in the compilation of "/home/users/chenrui17/github/mmcv/mmcv/ops/csrc/pytorch/cuda/carafe_cuda.cu".

Additional information

no

[grimoire]

#2608 Please have a try.

[chenrui17]

#2608 Please have a try.

Problem solved, Thanks a lot