RTX 5090 Failure Run

[azhuantou]

Hello, I have successfully installed tinycudann-2.0 on the 5090 GPU, but the code running failed. Has anyone run this on the 5090? It seems that the error message "You need C++17 to compile PyTorch" is the key point. But I still don't know how to solve this problem.

CUDA 12.8 Python 3.12 Pytorch 2.8.0

RuntimeError: Error building extension 'grid_encoder': [1/3] /usr/local/cuda-12.8/bin/nvcc --generate-dependencies-with-compile --dependency-output gridencoder.cuda.o.d -DTORCH_EXTENSION_NAME=grid_encoder -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1018" -isystem /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include -isystem /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda-12.8/include -isystem /home/lcd/anaconda3/envs/splat/include/python3.12 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS_ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_120,code=compute_120 -gencode=arch=compute_120,code=sm_120 --compiler-options '-fPIC' -O3 -std=c++14 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -c /mnt/largedisk/lcd/code/nest-splatting/gridencoder/src/gridencoder.cu -o gridencoder.cuda.o FAILED: [code=1] gridencoder.cuda.o /usr/local/cuda-12.8/bin/nvcc --generate-dependencies-with-compile --dependency-output gridencoder.cuda.o.d -DTORCH_EXTENSION_NAME=grid_encoder -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1018" -isystem /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include -isystem /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda-12.8/include -isystem /home/lcd/anaconda3/envs/splat/include/python3.12 -D__CUDA_NO_HALF_OPERATORS -D__CUDA_NO_HALF_CONVERSIONS -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_120,code=compute_120 -gencode=arch=compute_120,code=sm_120 --compiler-options '-fPIC' -O3 -std=c++14 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -c /mnt/largedisk/lcd/code/nest-splatting/gridencoder/src/gridencoder.cu -o gridencoder.cuda.o In file included from /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include/c10/util/CallOnce.h:4, from /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include/ATen/Context.h:25, from /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include/ATen/cuda/CUDAContext.h:6, from /mnt/largedisk/lcd/code/nest-splatting/gridencoder/src/gridencoder.cu:5: /home/lcd/anaconda3/envs/splat/lib/python3.12/site-packages/torch/include/c10/util/C++17.h:24:2: error: #error You need C++17 to compile PyTorch 24 | #error You need C++17 to compile PyTorch