error: identifier "__hmax" is undefined

[MrHandsomeljn]

Hello, I have a lots of compiling errors
First of all, my system is

• Ubuntu20.04
• cuda_11.3.r11.3/compiler.30059648_0【cuda-nvcc in conda environment】
• gcc&g++: 8.4.0 (Ubuntu 8.4.0-3ubuntu2)
• python 3.10 in conda environment
• pytorch1.12.1 py3.10_cuda11.3_cudnn8.3.2_0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch

I found a error while install with "pip install git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch".
The compiling begins well like:
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple Collecting git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch Cloning https://github.com/NVlabs/tiny-cuda-nn/ to /tmp/pip-req-build-10mg9n7j Running command git clone --filter=blob:none --quiet https://github.com/NVlabs/tiny-cuda-nn/ /tmp/pip-req-build-10mg9n7j Resolved https://github.com/NVlabs/tiny-cuda-nn/ to commit daf9628c8d4500aee29935e15cff58a359277ad0 Running command git submodule update --init --recursive -q Preparing metadata (setup.py): started Preparing metadata (setup.py): finished with status 'done' Building wheels for collected packages: tinycudann Building wheel for tinycudann (setup.py): started

Then errors:

` × python setup.py bdist_wheel did not run successfully.
│ exit code: 1
╰─> [2146 lines of output]
/tmp/pip-req-build-10mg9n7j/bindings/torch/setup.py:5: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
from pkg_resources import parse_version
Building PyTorch extension for tiny-cuda-nn version 1.7
Obtained compute capabilities [86] from environment variable TCNN_CUDA_ARCHITECTURES
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Wed_Jun__9_20:32:38_PDT_2021
Cuda compilation tools, release 11.3, V11.3.122
Build cuda_11.3.r11.3/compiler.30059648_0
Detected CUDA version 11.3
Targeting C++ standard 17
running bdist_wheel
running build
running build_py
creating build
creating build/lib.linux-x86_64-cpython-310
creating build/lib.linux-x86_64-cpython-310/tinycudann
copying tinycudann/init.py -> build/lib.linux-x86_64-cpython-310/tinycudann
copying tinycudann/modules.py -> build/lib.linux-x86_64-cpython-310/tinycudann
running egg_info
creating tinycudann.egg-info
writing tinycudann.egg-info/PKG-INFO
writing dependency_links to tinycudann.egg-info/dependency_links.txt
writing top-level names to tinycudann.egg-info/top_level.txt
writing manifest file 'tinycudann.egg-info/SOURCES.txt'
reading manifest file 'tinycudann.egg-info/SOURCES.txt'
writing manifest file 'tinycudann.egg-info/SOURCES.txt'
copying tinycudann/bindings.cpp -> build/lib.linux-x86_64-cpython-310/tinycudann
running build_ext
building 'tinycudann_bindings.86_C' extension
creating /tmp/pip-req-build-10mg9n7j/bindings/torch/dependencies
creating /tmp/pip-req-build-10mg9n7j/bindings/torch/dependencies/fmt
creating /tmp/pip-req-build-10mg9n7j/bindings/torch/dependencies/fmt/src
creating /tmp/pip-req-build-10mg9n7j/bindings/torch/src
creating /tmp/pip-req-build-10mg9n7j/bindings/torch/build/temp.linux-x86_64-cpython-310
creating /tmp/pip-req-build-10mg9n7j/bindings/torch/build/temp.linux-x86_64-cpython-310/tinycudann
Emitting ninja build file /tmp/pip-req-build-10mg9n7j/bindings/torch/build/temp.linux-x86_64-cpython-310/build.ninja...
Compiling objects...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/10] c++ -MMD -MF /tmp/pip-req-build-10mg9n7j/bindings/torch/dependencies/fmt/src/os.o.d -pthread -B /home/ljn/data/miniconda3/envs/CusNeRF/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall -fPIC -O2 -isystem /home/ljn/data/miniconda3/envs/CusNeRF/include -fPIC -O2 -isystem /home/ljn/data/miniconda3/envs/CusNeRF/include -fPIC -I/tmp/pip-req-build-10mg9n7j/include -I/tmp/pip-req-build-10mg9n7j/dependencies -I/tmp/pip-req-build-10mg9n7j/dependencies/cutlass/include -I/tmp/pip-req-build-10mg9n7j/dependencies/cutlass/tools/util/include -I/tmp/pip-req-build-10mg9n7j/dependencies/fmt/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/TH -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/THC -I/home/ljn/data/miniconda3/envs/CusNeRF/include -I/home/ljn/data/miniconda3/envs/CusNeRF/include/python3.10 -c -c /tmp/pip-req-build-10mg9n7j/dependencies/fmt/src/os.cc -o /tmp/pip-req-build-10mg9n7j/bindings/torch/dependencies/fmt/src/os.o -std=c++17 -DTCNN_PARAMS_UNALIGNED -DTCNN_MIN_GPU_ARCH=86 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="gcc"' '-DPYBIND11_STDLIB="libstdcpp"' '-DPYBIND11_BUILD_ABI="cxxabi1011"' -DTORCH_EXTENSION_NAME=86_C -D_GLIBCXX_USE_CXX11_ABI=0
[2/10] /home/ljn/data/miniconda3/envs/CusNeRF/bin/nvcc -I/tmp/pip-req-build-10mg9n7j/include -I/tmp/pip-req-build-10mg9n7j/dependencies -I/tmp/pip-req-build-10mg9n7j/dependencies/cutlass/include -I/tmp/pip-req-build-10mg9n7j/dependencies/cutlass/tools/util/include -I/tmp/pip-req-build-10mg9n7j/dependencies/fmt/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/TH -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/THC -I/home/ljn/data/miniconda3/envs/CusNeRF/include -I/home/ljn/data/miniconda3/envs/CusNeRF/include/python3.10 -c -c /tmp/pip-req-build-10mg9n7j/src/object.cu -o /tmp/pip-req-build-10mg9n7j/bindings/torch/src/object.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'"'"'' -std=c++17 --extended-lambda --expt-relaxed-constexpr -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -Xcompiler=-Wno-float-conversion -Xcompiler=-fno-strict-aliasing -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 -DTCNN_PARAMS_UNALIGNED -DTCNN_MIN_GPU_ARCH=86 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="gcc"' '-DPYBIND11_STDLIB="libstdcpp"' '-DPYBIND11_BUILD_ABI="cxxabi1011"' -DTORCH_EXTENSION_NAME=86_C -D_GLIBCXX_USE_CXX11_ABI=0
FAILED: /tmp/pip-req-build-10mg9n7j/bindings/torch/src/object.o
/home/ljn/data/miniconda3/envs/CusNeRF/bin/nvcc -I/tmp/pip-req-build-10mg9n7j/include -I/tmp/pip-req-build-10mg9n7j/dependencies -I/tmp/pip-req-build-10mg9n7j/dependencies/cutlass/include -I/tmp/pip-req-build-10mg9n7j/dependencies/cutlass/tools/util/include -I/tmp/pip-req-build-10mg9n7j/dependencies/fmt/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/TH -I/home/ljn/data/miniconda3/envs/CusNeRF/lib/python3.10/site-packages/torch/include/THC -I/home/ljn/data/miniconda3/envs/CusNeRF/include -I/home/ljn/data/miniconda3/envs/CusNeRF/include/python3.10 -c -c /tmp/pip-req-build-10mg9n7j/src/object.cu -o /tmp/pip-req-build-10mg9n7j/bindings/torch/src/object.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'"'"'' -std=c++17 --extended-lambda --expt-relaxed-constexpr -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -Xcompiler=-Wno-float-conversion -Xcompiler=-fno-strict-aliasing -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 -DTCNN_PARAMS_UNALIGNED -DTCNN_MIN_GPU_ARCH=86 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=_86_C -D_GLIBCXX_USE_CXX11_ABI=0
/tmp/pip-req-build-10mg9n7j/include/tiny-cuda-nn/common_device.h(94): error: identifier "__hmax" is undefined

  /tmp/pip-req-build-10mg9n7j/dependencies/fmt/include/fmt/core.h(288): warning: unrecognized GCC pragma
  
  /tmp/pip-req-build-10mg9n7j/include/tiny-cuda-nn/cuda_graph.h(137): error: identifier "cudaGraphExecUpdateResult" is undefined
  
  /tmp/pip-req-build-10mg9n7j/include/tiny-cuda-nn/cuda_graph.h(139): error: identifier "cudaGraphExecUpdate" is undefined
  
  /tmp/pip-req-build-10mg9n7j/include/tiny-cuda-nn/cuda_graph.h(142): error: identifier "cudaGraphExecUpdateSuccess" is undefined
  
  /tmp/pip-req-build-10mg9n7j/include/tiny-cuda-nn/gpu_memory.h(689): error: identifier "CUmemGenericAllocationHandle" is undefined
  
  /usr/include/c++/8/bits/ptr_traits.h(114): error: static assertion failed with "pointer type defines element_type or is like SomePointer<T, Args>"
            detected during:`

Pages of that error. The same if i compile the downloaded code. It's just begin with "__hmax is underfinded" too.

I tried gcc/g++9, don't work.
If there's some operational mistakes, please let me know. Best regards.

[MrHandsomeljn]

done!
check your path, like this:

• CUDA_HOME=/usr/local/cuda-11.8
• GCC_HOME=/usr/bin/gcc-9
• LD_LIBRARY_PATH=${GCC_HOME}/lib64:${CUDA_HOME}/lib64:${CUDA_HOME}/extras/CUPTI/lib64

NOTE: cuda-nvcc installed by conda is not supported
Recommend writing it in readme. Waste of time