Ubuntu 22.04 Launches but hangs on first generate

[drago87]

i have followed this quide How to use Stable Diffusion XL locally with AMD ROCm. With AUTOMATIC1111 WebUI and ComfyUI on Linux.
the only difference i did was
pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm5.7 (insted of the 5.6)
and
pip3 install -r requirements_versions.txt (instead of requirements.txt)

The problem i have is that when i start it it seam to start fine but as soon as i try generate a image the GPU spikes to 99% and 53% VRAM nothing more happens. no error (that i can find) and stays there untill i close the terminal.
i can not stop it with ctrl+c

I have a RX 5700 XT 8GB

(base) drago87@drago87-PC:~/stable-diffusion-webui$ conda activate sd
(sd) drago87@drago87-PC:~/stable-diffusion-webui$ bash launch.sh
Python 3.10.6 (main, Oct 24 2022, 16:07:47) [GCC 11.2.0]
Version: v1.6.0-2-g4afaaf8a
Commit hash: 4afaaf8a020c1df457bcf7250cb1c7f609699fa7
Launching Web UI with arguments: --listen --enable-insecure-extension-access --opt-sdp-attention
no module 'xformers'. Processing without...
no module 'xformers'. Processing without...
No module 'xformers'. Proceeding without it.
Loading weights [6ce0161689] from /home/drago87/stable-diffusion-webui/models/Stable-diffusion/v1-5-pruned-emaonly.safetensors
Running on local URL:  http://0.0.0.0:7860

To create a public link, set `share=True` in `launch()`.
Startup time: 18.0s (prepare environment: 2.0s, import torch: 7.4s, import gradio: 2.0s, setup paths: 2.2s, initialize shared: 0.1s, other imports: 3.3s, load scripts: 0.2s, create ui: 0.4s, gradio launch: 0.2s).
Creating model from config: /home/drago87/stable-diffusion-webui/configs/v1-inference.yaml
Applying attention optimization: sdp... done.

[syboot]

hi drago87
Launching Web UI with arguments: --listen --use-cpu
use CPU as torch device for specified modules

1,Remove the other parameters
2,add --use-cpu
3.Test out the basic features，Troubleshoot the issue if it is a GPU driver。

[drago87]

I have tried that now and this is what i get when i launch it

Python 3.10.6 (main, Oct 24 2022, 16:07:47) [GCC 11.2.0]
Version: v1.6.0-2-g4afaaf8a
Commit hash: 4afaaf8a020c1df457bcf7250cb1c7f609699fa7
Launching Web UI with arguments: --listen --use-cpu ALL
no module 'xformers'. Processing without...
no module 'xformers'. Processing without...
No module 'xformers'. Proceeding without it.
Warning: caught exception 'Expected a cuda device, but got: cpu', memory monitor disabled
Loading weights [f228f2a8a2] from /home/drago87/stable-diffusion-webui/models/Stable-diffusion/realcartoon3d_v9.safetensors
Running on local URL:  http://0.0.0.0:7861

To create a public link, set `share=True` in `launch()`.
Startup time: 6.3s (prepare environment: 1.5s, import torch: 2.4s, import gradio: 0.5s, setup paths: 0.5s, other imports: 0.3s, load scripts: 0.3s, create ui: 0.5s, gradio launch: 0.1s).
Creating model from config: /home/drago87/stable-diffusion-webui/configs/v1-inference.yaml
Applying attention optimization: Doggettx... done.
loading stable diffusion model: RuntimeError
Traceback (most recent call last):
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 973, in _bootstrap
    self._bootstrap_inner()
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
    self.run()
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 953, in run
    self._target(*self._args, **self._kwargs)
  File "/home/drago87/stable-diffusion-webui/modules/initialize.py", line 147, in load_model
    shared.sd_model  # noqa: B018
  File "/home/drago87/stable-diffusion-webui/modules/shared_items.py", line 110, in sd_model
    return modules.sd_models.model_data.get_sd_model()
  File "/home/drago87/stable-diffusion-webui/modules/sd_models.py", line 499, in get_sd_model
    load_model()
  File "/home/drago87/stable-diffusion-webui/modules/sd_models.py", line 649, in load_model
    sd_model.cond_stage_model_empty_prompt = get_empty_cond(sd_model)
  File "/home/drago87/stable-diffusion-webui/modules/sd_models.py", line 537, in get_empty_cond
    return sd_model.cond_stage_model([""])
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1510, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1519, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/drago87/stable-diffusion-webui/modules/sd_hijack_clip.py", line 234, in forward
    z = self.process_tokens(tokens, multipliers)
  File "/home/drago87/stable-diffusion-webui/modules/sd_hijack_clip.py", line 273, in process_tokens
    z = self.encode_with_transformers(tokens)
  File "/home/drago87/stable-diffusion-webui/modules/sd_hijack_clip.py", line 326, in encode_with_transformers
    outputs = self.wrapped.transformer(input_ids=tokens, output_hidden_states=-opts.CLIP_stop_at_last_layers)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1510, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1519, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/transformers/models/clip/modeling_clip.py", line 822, in forward
    return self.text_model(
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1510, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1519, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/transformers/models/clip/modeling_clip.py", line 740, in forward
    encoder_outputs = self.encoder(
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1510, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1519, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/transformers/models/clip/modeling_clip.py", line 654, in forward
    layer_outputs = encoder_layer(
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1510, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1519, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/transformers/models/clip/modeling_clip.py", line 382, in forward
    hidden_states = self.layer_norm1(hidden_states)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1510, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1519, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/drago87/stable-diffusion-webui/extensions-builtin/Lora/networks.py", line 474, in network_LayerNorm_forward
    return originals.LayerNorm_forward(self, input)
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/modules/normalization.py", line 196, in forward
    return F.layer_norm(
  File "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/torch/nn/functional.py", line 2548, in layer_norm
    return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'


Stable diffusion model failed to load

the GUI launches but when i try to generate a picture it take 1 sec and then it cancel it and i get this in the terminal

---
when i press ctrl+c i get this

Interrupted with signal 2 in <frame at 0xc3f1200, file '/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py', line 324, code wait>

# Thread: asyncio_0(140049295578688)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 973, in _bootstrap
  self._bootstrap_inner()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
  self.run()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 953, in run
  self._target(*self._args, **self._kwargs)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/concurrent/futures/thread.py", line 81, in _worker
  work_item = work_queue.get(block=True)

# Thread: AnyIO worker thread(140047743690304)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 973, in _bootstrap
  self._bootstrap_inner()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
  self.run()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 797, in run
  item = self.queue.get()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/queue.py", line 171, in get
  self.not_empty.wait()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 320, in wait
  waiter.acquire()

# Thread: AnyIO worker thread(140048960837184)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 973, in _bootstrap
  self._bootstrap_inner()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
  self.run()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 797, in run
  item = self.queue.get()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/queue.py", line 171, in get
  self.not_empty.wait()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 320, in wait
  waiter.acquire()

# Thread: Thread-4 (run)(140048983914048)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 973, in _bootstrap
  self._bootstrap_inner()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
  self.run()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 953, in run
  self._target(*self._args, **self._kwargs)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/site-packages/uvicorn/server.py", line 61, in run
  return asyncio.run(self.serve(sockets=sockets))
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/asyncio/runners.py", line 44, in run
  return loop.run_until_complete(main)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/asyncio/base_events.py", line 633, in run_until_complete
  self.run_forever()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/asyncio/base_events.py", line 600, in run_forever
  self._run_once()
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/asyncio/base_events.py", line 1860, in _run_once
  event_list = self._selector.select(timeout)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/selectors.py", line 469, in select
  fd_event_list = self._selector.poll(timeout, max_ev)

# Thread: MainThread(140061575578688)
File: "/home/drago87/stable-diffusion-webui/launch.py", line 48, in <module>
  main()
File: "/home/drago87/stable-diffusion-webui/launch.py", line 44, in main
  start()
File: "/home/drago87/stable-diffusion-webui/modules/launch_utils.py", line 436, in start
  webui.webui()
File: "/home/drago87/stable-diffusion-webui/webui.py", line 126, in webui
  server_command = shared.state.wait_for_server_command(timeout=5)
File: "/home/drago87/stable-diffusion-webui/modules/shared_state.py", line 62, in wait_for_server_command
  if self._server_command_signal.wait(timeout):
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 607, in wait
  signaled = self._cond.wait(timeout)
File: "/home/drago87/miniconda3/envs/sd/lib/python3.10/threading.py", line 324, in wait
  gotit = waiter.acquire(True, timeout)
File: "/home/drago87/stable-diffusion-webui/modules/initialize_util.py", line 156, in sigint_handler
  dumpstacks()
File: "/home/drago87/stable-diffusion-webui/modules/initialize_util.py", line 143, in dumpstacks
  for filename, lineno, name, line in traceback.extract_stack(stack):

[syboot]

Launching Web UI with arguments: --listen --skip-torch-cuda-test --precision full --no-half
Add these parameters try it. Is rocm5.6 also not working?

[drago87]

Running with those argument made no difference from the first run here is the Terminal

Python 3.10.6 (main, Oct 24 2022, 16:07:47) [GCC 11.2.0]
Version: v1.6.0-2-g4afaaf8a
Commit hash: 4afaaf8a020c1df457bcf7250cb1c7f609699fa7
Launching Web UI with arguments: --listen --skip-torch-cuda-test --precision full --no-half
no module 'xformers'. Processing without...
no module 'xformers'. Processing without...
No module 'xformers'. Proceeding without it.
Loading weights [f228f2a8a2] from /home/drago87/stable-diffusion-webui/models/Stable-diffusion/realcartoon3d_v9.safetensors
Running on local URL:  http://0.0.0.0:7861

To create a public link, set `share=True` in `launch()`.
Startup time: 4.7s (import torch: 2.4s, import gradio: 0.5s, setup paths: 0.5s, other imports: 0.3s, load scripts: 0.3s, create ui: 0.5s, gradio launch: 0.1s).
Creating model from config: /home/drago87/stable-diffusion-webui/configs/v1-inference.yaml
Applying attention optimization: Doggettx... done.

have not tried the rocm5.6 what is the best way to change it?

[syboot]

pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.6

[drago87]

after downgrading to 5.6 i'm getting this

Python 3.10.6 (main, Oct 24 2022, 16:07:47) [GCC 11.2.0]
Version: v1.6.0-2-g4afaaf8a
Commit hash: 4afaaf8a020c1df457bcf7250cb1c7f609699fa7
Launching Web UI with arguments: --listen --skip-torch-cuda-test --precision full --no-half
no module 'xformers'. Processing without...
no module 'xformers'. Processing without...
No module 'xformers'. Proceeding without it.
Loading weights [f228f2a8a2] from /home/drago87/stable-diffusion-webui/models/Stable-diffusion/realcartoon3d_v9.safetensors
Running on local URL:  http://0.0.0.0:7861

To create a public link, set `share=True` in `launch()`.
Startup time: 5.7s (import torch: 3.2s, import gradio: 0.5s, setup paths: 0.5s, other imports: 0.4s, load scripts: 0.3s, create ui: 0.5s, gradio launch: 0.1s).
Creating model from config: /home/drago87/stable-diffusion-webui/configs/v1-inference.yaml
Applying attention optimization: Doggettx... done.
Memory access fault by GPU node-1 (Agent handle: 0x7e14c50) on address 0x7ff8b43be000. Reason: Page not present or supervisor privilege.
launch.sh: line 9: 117037 Aborted                 (core dumped) python3 launch.py --listen --skip-torch-cuda-test --precision full --no-half

[syboot]

Try it with docker
install docker on OS
I'm also using docker run it

1.docker pull rocm/pytorch:latest
doc: https://rocmdocs.amd.com/en/latest/how_to/pytorch_install/pytorch_install.html
2.docker run -it --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --device=/dev/kfd --device=/dev/dri --group-add video --ipc=host --shm-size 8G rocm/pytorch:latest

[drago87]

tried installing docker with this Install using the apt repository
and ran the command you gave me and this is what i got

docker pull rocm/pytorch:latest
permission denied while trying to connect to the Docker daemon socket at unix:///var/run/docker.sock: Post "http://%2Fvar%2Frun%2Fdocker.sock/v1.24/images/create?fromImage=rocm%2Fpytorch&tag=latest": dial unix /var/run/docker.sock: connect: permission denied
(base) drago87@drago87-PC:~$ sudo docker pull rocm/pytorch:latest
latest: Pulling from rocm/pytorch
edaedc954fb5: Pulling fs layer 
fda2994f7c22: Pulling fs layer 
a91974e2797b: Pulling fs layer 
76e4290a2a40: Waiting 
213cac4e843b: Waiting 
87a4dc57ffbb: Waiting 
2f88b0371440: Pulling fs layer 
6c4e02617376: Waiting 
2b20318d27fa: Waiting 
3dfb34601767: Pulling fs layer 
f31f4760e599: Waiting 
0e552d527f1f: Pulling fs layer 
1dad240711a9: Waiting 
80dded4e0c20: Waiting 
8591724c8ad1: Waiting 
5f958ceac03e: Pulling fs layer 
df643497f446: Pulling fs layer 
4116d5dbdd40: Pulling fs layer 
d555b170c0a1: Waiting 
97af2371e0e0: Waiting 
1bda41165ba1: Waiting 
39d5d06f0bb8: Pull complete 
0cfd6770bd0d: Pull complete 
9c4cb18a13be: Pull complete 
ebb09f80ca9d: Pull complete 
c466916b06f5: Pull complete 
7159128bc173: Pull complete 
82e85c0208f4: Pull complete 
806df43b4aff: Pull complete 
d11147f23fa2: Pull complete 
4ab2cec8a410: Pull complete 
092208ab6d04: Pull complete 
c98315a72a6a: Pull complete 
681ddfa7aeea: Pull complete 
2d9a7e99a879: Pull complete 
e40025d4cdbc: Pull complete 
84f149b8473e: Pull complete 
c10a2675e244: Pull complete 
555dcf06087a: Pull complete 
7b598d8ea07e: Pull complete 
6dd55ea333c2: Pull complete 
3cec5d3f9905: Pull complete 
8e61eb7bb401: Pull complete 
9c331cb4feac: Pull complete 
Digest: sha256:56ea92985c9445fe524fcb2a6b657994599a5b81b69ed36559c3eb2724cee864
Status: Downloaded newer image for rocm/pytorch:latest
docker.io/rocm/pytorch:latest
(base) drago87@drago87-PC:~$ docker run -it --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --device=/dev/kfd --device=/dev/dri --group-add video --ipc=host --shm-size 8G rocm/pytorch:latest
docker: unknown server OS: .
See 'docker run --help'.