Commit Graph

32 Commits

Author SHA1 Message Date
Marc-Andre Ferland
86da27a7a1 Moved wait outside lock and now returns false on failure. 2022-10-28 22:52:00 -04:00
Marc-Andre Ferland
fc2a6567da Moved import before use of runtime.thread_data.device 2022-10-28 22:51:04 -04:00
cmdr2
a8c16e39b8 Support custom VAE files; Use vae-ft-mse-840000-ema-pruned as the default VAE, which can be overridden by putting a .vae.pt file inside models/stable-diffusion with the same name as the ckpt model file. The UI / System Settings allows setting the default VAE model to use 2022-10-28 20:06:44 +05:30
Marc-Andre Ferland
26562e445f Set online after preload. Move ident to include in if check. 2022-10-28 04:09:34 -04:00
Marc-Andre Ferland
22a11769fa Enable preload on cpu when no other devices are alive. 2022-10-27 21:57:50 -04:00
Marc-Andre Ferland
7f4786f9dd Wait until device is fully ready before proceding. 2022-10-27 20:27:21 -04:00
cmdr2
0dfaf9159d Put back the check to only preload on GPU 2022-10-28 00:04:33 +05:30
cmdr2
389e3397ec Preload the model even in the CPU mode 2022-10-27 23:17:41 +05:30
Marc-Andre Ferland
ae40b6ba8c Missed a is_alive check in the conversion. 2022-10-25 03:00:50 -04:00
Marc-Andre Ferland
364e364429 Added get_cached_task to replace task_cache.tryGet in server.py
Now updated cache TTL on /stream and temp images endpoints.
Keep images alive longer when browser keeps reading the endpoints.
2022-10-22 13:52:13 -04:00
Marc-Andre Ferland
3b5f96a133 Fixed stopping tasks and more cleaning. 2022-10-21 22:45:19 -04:00
Marc-Andre Ferland
56ed4fe6f2 Fix VisualStudio Type Warning. 2022-10-21 01:30:49 -04:00
Marc-Andre Ferland
849d1d7ebd Merge branch 'beta' of https://github.com/cmdr2/stable-diffusion-ui.git into multi-gpu
# Conflicts:
#	ui/media/js/main.js
#	ui/sd_internal/runtime.py
#	ui/server.py
2022-10-20 20:08:23 -04:00
cmdr2
090dfff730 Refactor the time delays into constants and mention the units 2022-10-20 17:22:01 +05:30
Marc-Andre Ferland
4e5ddca3bd Display the failure detail when there is one at that step.
Was checking the json object, not the server response.
2022-10-19 05:10:37 -04:00
Marc-Andre Ferland
3bdc90451a Dont preload on cpu. 2022-10-19 04:34:54 -04:00
Marc-Andre Ferland
21afe077d7 Removed Cpu from the devices allowed to run GFPGANer.
Added clear error for the user.
2022-10-19 03:02:26 -04:00
Marc-Andre Ferland
6098b196dc Text header, comments and better validations. 2022-10-18 23:58:55 -04:00
Marc-Andre Ferland
fcdb086daf Fixed is_alive to work with devices that can change name after init. 2022-10-18 20:33:37 -04:00
Marc-Andre Ferland
5e461e9b6b Fixed is_alive with render_threads that can update the device name after starting. 2022-10-18 13:21:15 -04:00
Marc-Andre Ferland
e24be913e5 Merge branch 'beta' into multi-gpu 2022-10-17 21:35:24 -04:00
cmdr2
cb82170187 Fix the bug where custom models weren't getting picked up 2022-10-17 21:22:05 +05:30
Marc-Andre Ferland
c92129ac63 Improved detection of missing cuda:0 and added warning to console about how to fix. 2022-10-17 03:32:23 -04:00
Marc-Andre Ferland
554b67a2f0 Fixing bug in is_alive. 2022-10-17 01:05:51 -04:00
Marc-Andre Ferland
012243a880 Process GPU tasks on CPU when there are no cuda devices at all. 2022-10-17 01:05:27 -04:00
Marc-Andre Ferland
d4a348a2b2 Process GFPGANer on cuda:0 when possible, otherwise use cpu. 2022-10-16 23:12:46 -04:00
Marc-Andre Ferland
41bfb96b6b Fixed bug in task_manager.is_alive and added way to check for first device. 2022-10-16 23:06:41 -04:00
Marc-Andre Ferland
7c72608e1c First draft for Multi-GPU support 2022-10-16 21:41:39 -04:00
Marc-Andre Ferland
cbdf03450d Added timeout to critical locking tasks with matching exception 2022-10-15 05:31:17 -04:00
Marc-Andre Ferland
8fdb1e7ec9 Improved locking and logging when cleaning old cached sessions. 2022-10-15 04:39:45 -04:00
Marc-Andre Ferland
d3b28c42e6 Better error handling with cache.put 2022-10-15 04:08:17 -04:00
Marc-Andre Ferland
7de699c7fa Moved a lot of code into task_manager.py 2022-10-15 03:28:20 -04:00