Closed rohitdhote111 closed 1 year ago
Your Cuda toolkit installation appears to be incomplete. cudart
is the Cuda runtime library.
@s-laine Hi, Thanks for the reply I am using docker container pytorch/pytorch:1.13.0-cuda11.6-cudnn8-devel so cuda toolkit is already installed. I also checked it with nvcc --verion command. Let me know if I am missing something
For some reason libcudart.so is not found by the linker on its library path, and I cannot think of why this would happen. Maybe the conda environment is somehow bad and doesn't have the correct environment variables for the linker? There's discussion about a similar issue here, although it's mostly about the main Cuda library libcuda.so.
solved in #115
FAILED: nvdiffrast_plugin_gl.so c++ common.o glutil.o rasterize_gl.o torch_bindings_gl.o torch_rasterize_gl.o -shared -lGL -lEGL -L/opt/conda/lib/python3.9/site-packages/torch/lib -lc10 -lc10_cuda -ltorch_cpu -ltorch_cuda_cu -ltorch_cuda_cpp -ltorch -ltorch_python -L/opt/conda/lib64 -lcudart -o nvdiffrast_plugin_gl.so /usr/bin/ld: cannot find -lcudart collect2: error: ld returned 1 exit status ninja: build stopped: subcommand failed.