You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I attempted to submit this as an issue, but the New Issue button gave only options related to issues in the web UI itself, saying " You think somethings is broken in the UI ". Given that my issue isn't in the UI but rather the execution, I instead believe my situation matches " Please ask and answer questions here. ", which took me to this Discussions page.
Now then. I was previously running Ubuntu 22.04, but ran into this same issue, so I decided to try Mint 21, but there has been no improvement from going through a fresh install. So for now, I'm on Mint 21. I'm running with an 8 GB RX 580.
Note that I'm running Natively, not in Docker. I then get the error:
stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/cuda/__init__.py:88: UserWarning: HIP initialization: Unexpected error from hipGetDeviceCount(). Did you run some cuda functions before calling NumHipDevices() that might have already set an error? Error 101: hipErrorInvalidDevice (Triggered internally at ../c10/hip/HIPFunctions.cpp:110.) return torch._C._cuda_getDeviceCount() > 0 Warning: caught exception 'Unexpected error from hipGetDeviceCount(). Did you run some cuda functions before calling NumHipDevices() that might have already set an error? Error 101: hipErrorInvalidDevice', memory monitor disabled
From there, I found the Issue page:
I started by changing line 129 of launch.py, so it now says commandline_args = os.environ.get('COMMANDLINE_ARGS', "--skip-torch-cuda-test"). This creates no change in behavior.
Then I went to the last comment in the Issue page, which describes a big list of things to install. However, when I try to install amdgpu-core, it says the package is not found.
What should I try next in order to get this to work? If the project was officially Nvidia-only, I would have just walked away and understood that restriction, but as a whole things are said to be AMD-compatible, so my hope is that with the right tweaking I'll be able to get it to work. Can anyone help guide me through finding how to get hipGetDeviceCount to provide the expected result?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
-
Hello,
I attempted to submit this as an issue, but the New Issue button gave only options related to issues in the web UI itself, saying " You think somethings is broken in the UI ". Given that my issue isn't in the UI but rather the execution, I instead believe my situation matches " Please ask and answer questions here. ", which took me to this Discussions page.
Now then. I was previously running Ubuntu 22.04, but ran into this same issue, so I decided to try Mint 21, but there has been no improvement from going through a fresh install. So for now, I'm on Mint 21. I'm running with an 8 GB RX 580.
I began by following the guide here: https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Install-and-Run-on-AMD-GPUs
Note that I'm running Natively, not in Docker. I then get the error:
stable-diffusion-webui/venv/lib/python3.10/site-packages/torch/cuda/__init__.py:88: UserWarning: HIP initialization: Unexpected error from hipGetDeviceCount(). Did you run some cuda functions before calling NumHipDevices() that might have already set an error? Error 101: hipErrorInvalidDevice (Triggered internally at ../c10/hip/HIPFunctions.cpp:110.) return torch._C._cuda_getDeviceCount() > 0 Warning: caught exception 'Unexpected error from hipGetDeviceCount(). Did you run some cuda functions before calling NumHipDevices() that might have already set an error? Error 101: hipErrorInvalidDevice', memory monitor disabled
From there, I found the Issue page:
#1122
I started by changing line 129 of launch.py, so it now says
commandline_args = os.environ.get('COMMANDLINE_ARGS', "--skip-torch-cuda-test")
. This creates no change in behavior.Then I went to the last comment in the Issue page, which describes a big list of things to install. However, when I try to install amdgpu-core, it says the package is not found.
What should I try next in order to get this to work? If the project was officially Nvidia-only, I would have just walked away and understood that restriction, but as a whole things are said to be AMD-compatible, so my hope is that with the right tweaking I'll be able to get it to work. Can anyone help guide me through finding how to get hipGetDeviceCount to provide the expected result?
Beta Was this translation helpful? Give feedback.
All reactions