search

CS1o / Stable-Diffusion-Info

Stable Diffusion Knowledge Base (Setups, Basics, Guides and more)
26 stars 0 forks source link

RX580 / ZLUDA / gfx803 / error No such file or directory for GPU arch : gfx803 #2

Open RenanOSX opened 4 weeks ago

RenanOSX commented 4 weeks ago

GPU: RX580 OS: Windows 11 ROCm: 5.7 (The same one as the tutorial for RX580)

Following the tutorial for RX580 "[AMD] Automatic1111 with ZLUDA" I can't make it work as I get the following error:

venv "C:\SD-Zluda\stable-diffusion-webui-amdgpu\venv\Scripts\Python.exe"
WARNING: ZLUDA works best with SD.Next. Please consider migrating to SD.Next.
Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug  1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)]
Version: v1.10.1-amd-2-g395ce8dc
Commit hash: 395ce8dc2cb01282d48074a89a5e6cb3da4b59ab
Using ZLUDA in C:\SD-Zluda\stable-diffusion-webui-amdgpu\.zluda
Skipping onnxruntime installation.
You are up to date with the most recent release.
no module 'xformers'. Processing without...
no module 'xformers'. Processing without...
No module 'xformers'. Proceeding without it.
C:\SD-Zluda\stable-diffusion-webui-amdgpu\venv\lib\site-packages\pytorch_lightning\utilities\distributed.py:258: LightningDeprecationWarning: `pytorch_lightning.utilities.distributed.rank_zero_only` has been deprecated in v1.8.1 and will be removed in v2.0.0. You can import it from `pytorch_lightning.utilities` instead.
  rank_zero_deprecation(
Launching Web UI with arguments: --use-zluda --update-check --skip-ort

rocBLAS error: Cannot read C:\Program Files\AMD\ROCm\5.7\bin\/rocblas/library/TensileLibrary.dat: No such file or directory for GPU arch : gfx803

rocBLAS error: Could not initialize Tensile host:
regex_error(error_backref): The expression contained an invalid back reference.

Looking at the C:\Program Files\AMD\ROCm\5.7\bin\/rocblas/library/ I don't have all the files for gfx803, the only files that I got for gfx803 was the "TensileLibrary_gfx803.co" and "Kernels.so-000-gfx803.hsaco" doesn't seem to have the "TensileLibrary_lazy_gfx803.dat" as the other ones

If necessary any extra info let me know! Thank you!

CS1o commented 4 weeks ago

Hey normaly those two were enough but what you can try is to use one of these two versions below from: Source

Your gpu can use this: https://github.com/likelovewant/ROCmLibs-for-gfx1103-AMD780M-APU/releases/download/v0.5.7/rocm.gfx803.optic.test.version.7z

or this: https://github.com/likelovewant/ROCmLibs-for-gfx1103-AMD780M-APU/releases/download/v0.5.7/rocblas.for.gfx803.override.with.vega10.7z

in both cases you need to replace the rocm library files again. best is to do this with a clean copy of the library folder. You also need to place the rocblas.dll from the zip, inside the C:\Program Files\AMD\ROCm\5.7\bin\rocblas folder.

A PC restart is always required after changing library files.

Let me know if it worked with these files.

RenanOSX commented 3 weeks ago

Sorry for the late reply, sadly after opening the browser tab and loading the model I got this error:

edit: After trying again I got the loading bar working and when got to 100% the same error appeared (I tried to download the amdhip64.dll but I got mostly the same error.

venv "C:\SD-Zluda\stable-diffusion-webui-amdgpu\venv\Scripts\Python.exe"
WARNING: ZLUDA works best with SD.Next. Please consider migrating to SD.Next.
Python 3.10.6 (tags/v3.10.6:9c7b4bd, Aug  1 2022, 21:53:49) [MSC v.1932 64 bit (AMD64)]
Version: v1.10.1-amd-2-g395ce8dc
Commit hash: 395ce8dc2cb01282d48074a89a5e6cb3da4b59ab
Using ZLUDA in C:\SD-Zluda\stable-diffusion-webui-amdgpu\.zluda
Skipping onnxruntime installation.
You are up to date with the most recent release.
no module 'xformers'. Processing without...
no module 'xformers'. Processing without...
No module 'xformers'. Proceeding without it.
C:\SD-Zluda\stable-diffusion-webui-amdgpu\venv\lib\site-packages\pytorch_lightning\utilities\distributed.py:258: LightningDeprecationWarning: `pytorch_lightning.utilities.distributed.rank_zero_only` has been deprecated in v1.8.1 and will be removed in v2.0.0. You can import it from `pytorch_lightning.utilities` instead.
  rank_zero_deprecation(
Launching Web UI with arguments: --use-zluda --update-check --skip-ort
Loading weights [6ce0161689] from C:\SD-Zluda\stable-diffusion-webui-amdgpu\models\Stable-diffusion\v1-5-pruned-emaonly.safetensors
Creating model from config: C:\SD-Zluda\stable-diffusion-webui-amdgpu\configs\v1-inference.yaml
C:\SD-Zluda\stable-diffusion-webui-amdgpu\venv\lib\site-packages\huggingface_hub\file_download.py:1150: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
  warnings.warn(
Running on local URL:  http://127.0.0.1:7860

To create a public link, set `share=True` in `launch()`.
Startup time: 16.0s (prepare environment: 21.1s, initialize shared: 1.5s, other imports: 0.8s, load scripts: 0.8s, create ui: 1.2s, gradio launch: 0.5s).
Applying attention optimization: Doggettx... done.
Exception Code: 0xC0000005
0x00007FFC28A18D72, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x3A8D72 byte(s), hipMemUnmap() + 0x3CE32 byte(s)
0x00007FFC28A2ED70, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x3BED70 byte(s), hipMemUnmap() + 0x52E30 byte(s)
0x00007FFC28A2ADA4, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x3BADA4 byte(s), hipMemUnmap() + 0x4EE64 byte(s)
0x00007FFC2878CAD8, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x11CAD8 byte(s), hipPeekAtLastError() + 0x8218 byte(s)
0x00007FFC28A18D82, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x3A8D82 byte(s), hipMemUnmap() + 0x3CE42 byte(s)
0x00007FFC28A96C6C, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x426C6C byte(s), hipMemUnmap() + 0xBAD2C byte(s)
0x00007FFC28A9A87C, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x42A87C byte(s), hipMemUnmap() + 0xBE93C byte(s)
0x00007FFC28A30B2A, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x3C0B2A byte(s), hipMemUnmap() + 0x54BEA byte(s)
0x00007FFC28A30C11, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x3C0C11 byte(s), hipMemUnmap() + 0x54CD1 byte(s)
0x00007FFC289DD857, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x36D857 byte(s), hipMemUnmap() + 0x1917 byte(s)
0x00007FFC289DF7FF, C:\Windows\SYSTEM32\amdhip64.dll(0x00007FFC28670000) + 0x36F7FF byte(s), hipMemUnmap() + 0x38BF byte(s)
0x00007FFCC1B253E0, C:\Windows\System32\KERNEL32.DLL(0x00007FFCC1B10000) + 0x153E0 byte(s), BaseThreadInitThunk() + 0x10 byte(s)
0x00007FFCC3CC485B, C:\Windows\SYSTEM32\ntdll.dll(0x00007FFCC3CC0000) + 0x485B byte(s), RtlUserThreadStart() + 0x2B byte(s)
CS1o commented 3 weeks ago

Make sure your on the latest AMD Adrenalin Driver and also try to delete the .zluda folder from the stable-diffusion-webui-amdgpu folder. Also make sure youve did the Extra step needed for RX580 (installing torch 2.2.1 inside the venv). Then relaunch the webui-user.bat

RenanOSX commented 1 week ago

Same error as #4