txt2img runs its process but fails to show any picture at 100%

[weddi-eddy]

Describe the bug txt2img runs its process but fails to show any picture at 100%. Displays a message that says: [W NNPACK.cpp:51] Could not initialize NNPACK! Reason: Unsupported hardware./it]

This message is displayed before the Total progress bar reaches 100%. The bar on the site reaches 100%, but the image on the site does not update from the blank image.

To Reproduce Steps to reproduce the behavior:

1. Type in the prompt for txt2img and click Generate.
2. Wait for it to generate. The first bar shows 100%. Then the NNPACK error and the next bar shows 100%.
3. No image is shown.

Expected behavior When the Total progress bar is reached, the image should appear.

Screenshots

Desktop (please complete the following information):

• OS: Linux
• Browser: Firefox
• Commit revision: "78d6aef30249530d8cc0b5023b2f8bf8c86f8446"

Additional context

[cluder]

what cpu/gpu hardware do you have ? (laptop/cpu model etc)

[valconius]

I have the same problem when using CPU with this command-line:

env CUDA_VISIBLE_DEVICES=-1 python3 launch.py --precision full --no-half --no-half-vae --use-cpu SD GFPGAN BSRGAN ESRGAN SCUNet CodeFormer --all --skip-torch-cuda-test

The error is:

[W NNPACK.cpp:53] Could not initialize NNPACK! Reason: Unsupported hardware.

That said, if I wait for a minute (well the CPU s/it is about 30s/it on euler so about the time of two iterations), then the image generates anyway in the output folder, but it does not appear in the webui. This is okay for my use case but I presume this would be annoying behavior for anyone actually relying on the CPU.

The OS is most recent Ubuntu (non-LTS), updated, installed with the following procedure and running under Qemu with KVM:

$ sudo apt-get -y install git
$ git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
$ cd stable-diffusion-webui/
$ sudo apt-get -y install python3.10-venv
$ python3 -m venv venv
$ source venv/bin/activate
$ pip install torch torchvision torchaudio
$ cd ~/stable-diffusion-webui/
$ source venv/bin/activate
$ <command listed above>

[Easterlings]

Same problem,nothing showed when process finished but images appear in the output folder.