Failed to create a PhysX CUDA Context Manager. Falling back to CPU.

[HIT-ZY]

运行python train的命令报一下错误：
[Warning] [carb.gym.plugin] Failed to create a PhysX CUDA Context Manager. Falling back to CPU.
Physics Engine: PhysX
Physics Device: cpu
GPU Pipeline: disabled
/home/eit-robot2/anaconda3/envs/humanoid/lib/python3.8/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_1670525541702/work/aten/src/ATen/native/TensorShape.cpp:3190.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
[Error] [carb.gym.plugin] Gym cuda error: invalid resource handle: ../../../source/plugins/carb/gym/impl/Gym/GymPhysX.cpp: 6137
[Error] [carb.gym.plugin] Must enable GPU pipeline to use state tensors
[Error] [carb.gym.plugin] Must enable GPU pipeline to use state tensors
Traceback (most recent call last):
File "scripts/train.py", line 43, in
train(args)
File "scripts/train.py", line 37, in train
env, env_cfg = task_registry.make_env(name=args.task, args=args)
File "/home/eit-robot2/Documents/humanoid-gym/humanoid/utils/task_registry.py", line 97, in make_env
env = task_class( cfg=env_cfg,
File "/home/eit-robot2/Documents/humanoid-gym/humanoid/envs/custom/humanoid_env.py", line 78, in init
super().init(cfg, sim_params, physics_engine, sim_device, headless)
File "/home/eit-robot2/Documents/humanoid-gym/humanoid/envs/base/legged_robot.py", line 80, in init
self._init_buffers()
File "/home/eit-robot2/Documents/humanoid-gym/humanoid/envs/base/legged_robot.py", line 452, in _init_buffers
self.base_euler_xyz = get_euler_xyz_tensor(self.base_quat)
File "/home/eit-robot2/Documents/humanoid-gym/humanoid/envs/base/legged_robot.py", line 51, in get_euler_xyz_tensor
r, p, w = get_euler_xyz(quat)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: CUDA driver error: an illegal memory access was encountered
感觉是没用上cuda，请大佬帮忙看看

[zlw21gxy]

Could you operate the IsaacGym example? Additionally, could you verify whether a GPU can be identified in PyTorch? A general recommendation: Examine these two aspects initially. Kindly furnish more testing outcomes if the issue persists.