| Age | Commit message (Collapse) | Author |
|---|
|
With the release candidates being out for both torch and vision,
webui should default to these over nightly for a more stable experience.
Stable release isn't excpected until October 4th:
https://dev-discuss.pytorch.org/c/release-announcements/27
|
|
Update torch for Navi 31 (7900 XT/XTX)
|
|
Navi 3 needs at least 5.5 which is only on the nightly chain, previous versions are no longer online (torch==2.1.0.dev-20230614+rocm5.5 torchvision==0.16.0.dev-20230614+rocm5.5 torchaudio==2.1.0.dev-20230614+rocm5.5).
so switch to nightly rocm5.6 without explicit versions this time
|
|
|
|
|
|
|
|
|
|
When passing `--novenv` flag to webui.sh it can skip venv.
Might be useful for installing in Docker since messing with venv in Docker might be a bit complicated.
Example usage:
`webui.sh --novenv`
Hope this gets approved and pushed into future versions of Web UI
|
|
|
|
WEBUI.SH Navi 3 Support
|
|
|
|
|
|
Navi 3 card now defaults to nightly torch to utilize rocm 5.5
for out-of-the-box support.
https://download.pytorch.org/whl/nightly/
While its not yet on the main pytorch "get started" site,
it still seems perfectly indexable via pip which is all we need.
With this I'm able to clone a fresh repo and immediately run ./webui.sh
on my 7900 XTX without any problems.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Added support for workarounds on Navi external GPU.
|
|
lspci detects VGA for main/integrated videocards and Display
for external videocards.
This commit should apply workarounds on computers with more than
one GPU. Useful for most laptops using weak iGPU and good dGPU.
Signed-off-by: Pablo Cholaky <waltercool@slash.cl>
|
|
Download ROCm for AMD GPU only if NVIDIA is not presented
|
|
Small fixes to prepare_tcmalloc for Debian/Ubuntu compatibility
|
|
|
|
|
|
|
|
- /usr/sbin (where ldconfig is usually located) is not typically on users' PATHs by default, so we set that variable before trying to run ldconfig.
- The libtcmalloc library is called libtcmalloc_minimal on Debian/Ubuntu systems. We now check whether libtcmalloc_minimal exists when running prepare_tcmalloc.
|
|
|
|
|
|
If only i proofread what i wrote
|
|
So apparently it works now? Before you would get "Pytorch cant use the GPU" but not anymore.
|
|
feat: use existing virtualenv if already active
|
|
|
|
|
|
|
|
|
|
Try using TCMalloc on Linux by default
|
|
|
|
When the user's home directory and username are inconsistent, an error message stating that the directory cannot be found will appear. Directly default the installation directory to the user's home directory
|
|
|
|
command
|
|
The old code tries to install the newest versions of pytorch, wich is currently 2.0. Forcing it to 1.13.1
|
|
|
|
Installation should not be assumed to be located within ~/home directory. User should be expected to install project anywhere and run the startup scripts while in stable-diffusion-webui directory.
See issue #8534
|
|
Installation should not be assumed to be located within ~/home directory. User should be expected to install project anywhere and run the startup scripts while in stable-diffusion-webui directory.
See issue #8534
|
|
|
|
This may be helpful for
https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/7028,
because we won't change working directory to the repo now, instead,
we will use any working directory. If we set working directory to
a path contains repo and the custom --data-dir, the problem in this
issue should be solved.
Howewer, this may be treated as an incompatible change if some code
assume the working directory is always the repo.
Also, there may be another solution that always let --data-dir be the
subdirectory of the repo, but personally I think this may not be what
we actually need.
As this issue mainly influent on Docker and I am not familiar with
.bat files, updating webui.bat is skipped.
webui.sh: source env from repo instead $PWD
|
|
|
