Isaac Gym is not able to use the GPU

Hello,
When I tried the python example joint_monkey.py I got the error message below.
It’s on Ubuntu 20.04, RTX 2070, no docker or something like that. The GPU works when using PyTorch. The smi results are also below.
How do I enable the GPU pipeline?
Then there are some subsequent problems but I assume those follow from this.
Grateful for your help!

(rlgpu) karl@omen:~$ python apps/isaacgym/python/examples/joint_monkey.py
Importing module ‘gym_37’ (/home/karl/apps/isaacgym/python/isaacgym/_bindings/linux-x86_64/gym_37.so)
Setting GYM_USD_PLUG_INFO_PATH to /home/karl/apps/isaacgym/python/isaacgym/_bindings/linux-x86_64/usd/plugInfo.json
WARNING: Forcing CPU pipeline.
Not connected to PVD
+++ Using GPU PhysX
Physics Engine: PhysX
Physics Device: cuda:0
GPU Pipeline: disabled
Loading asset ‘mjcf/nv_humanoid.xml’ from ‘…/…/assets’
[Error] [carb.gym.plugin] *** Failed to load ‘…/…/assets/mjcf/nv_humanoid.xml’
[Error] [carb.gym.plugin] *** Failed to load ‘mjcf/nv_humanoid.xml’ from ‘…/…/assets’
Creating 36 environments
Traceback (most recent call last):
File “apps/isaacgym/python/examples/joint_monkey.py”, line 214, in
print(“Animating DOF %d (’%s’)” % (current_dof, dof_names[current_dof]))
IndexError: list index out of range

(rlgpu) karl@omen:~$ nvidia-smi
Tue Mar 22 23:48:04 2022
±----------------------------------------------------------------------------+
| NVIDIA-SMI 470.63.01 Driver Version: 470.63.01 CUDA Version: 11.4 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce … Off | 00000000:01:00.0 On | N/A |
| 30% 34C P8 23W / 210W | 1065MiB / 7948MiB | 2% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| 0 N/A N/A 899 G /usr/lib/xorg/Xorg 102MiB |
| 0 N/A N/A 1599 G /usr/lib/xorg/Xorg 512MiB |
| 0 N/A N/A 1736 G /usr/bin/gnome-shell 163MiB |
| 0 N/A N/A 13774 G …750682524097145037,131072 179MiB |
| 0 N/A N/A 27981 G …AAAAAAAAA= --shared-files 94MiB |
±----------------------------------------------------------------------------+

Hi @karlengblom,

The examples expect to be run from inside their own directory. The error message above refers to the code not being able to find the asset files it’s looking for in …/…/assets.

Try this:

cd apps/isaacgym/python/examples/
python joint_monkey.py

Note that many of the examples don’t enable the GPU pipeline by default. Just seeing GPU Pipeline: disabled does not indicate an issue with the GPU.

Take care,
-Gav

1 Like

Ok, er, sorry for that. It works now. Thanks for replying. Also thanks for letting us preview this very cool library.

1 Like