Segmentation fault of OptiXContext

[sungmin9939]

Hi

I have encountered segmentation fault while testing your work with this command python train_gshelltet_deepfashion.py --config configs/deepfashion_mc_256.json --index 0 --trainset_path data/deepfashion3d_training_data/ --testset_path data/deepfashion3d_test_data/ --o ./result

DatasetNERF: 72 images with shape [1024, 1024] DatasetNERF: 72 images with shape [1024, 1024] DatasetNERF: 200 images with shape [1024, 1024] /opt/conda/lib/python3.10/site-packages/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1695392020201/work/aten/src/ATen/native/TensorShape.cpp:3526.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] Cuda path /usr/local/cuda Segmentation fault (core dumped)

the exact location of code that yields segmentation fault is at OptiXStateWrapper of render/optixutils/ops.py

class OptiXContext: def init(self): print("Cuda path", torch.utils.cpp_extension.CUDA_HOME) self.cpp_wrapper = _plugin.OptiXStateWrapper(os.path.dirname(file), torch.utils.cpp_extension.CUDA_HOME)

and here is information of my environment

pytorch3d 0.7.5 torch 2.1.0 torchvision 0.16.0 Cuda: 11.8 docker image: pytorch/pytorch:2.1.0-cuda11.8-cudnn8-devel gcc version: 9.4.0

Please let me know how to solve this issue

[lzzcd001]

Without further detailed error messages, I can only tell that there seems to be some compatibility issue -- try check your GPU driver versions.

Related issue: https://github.com/NVlabs/nvdiffrecmc/issues/13

[sungmin9939]

my GPU is RTX 3090 and the driver version is 535.146.02 could relatively higher driver version be a problem? it would be thankful if you can let me know name of your gpu hardware and driver version you used to implement gshell

[lzzcd001]

Hey and sorry for the late response. Hope you have already solved the problem but here is the driver version in one of my working environments: 545.23.08