Commit Graph

286 Commits (3a8ef750687d8d0d4e2575bd16f796b8dfc15cdb)

Author SHA1 Message Date
Disty0 dfebc909eb Disable cuDNN benchmark on ROCm and add cudnn_benchmark_limit option 2025-05-08 13:27:06 +03:00
Disty0 90f887ac4a Add dim checks to ck flash atten and fix dim check on dyn atten 2025-03-25 03:50:21 +03:00
Seunghoon Lee 0c890b50e0
proper zluda detection 2025-03-20 23:03:23 +09:00
Disty0 1e0f512ccb ROCm disable FP16 for gfx1102 2025-03-19 15:42:36 +03:00
Disty0 878cab085f Reverse the sdpa hijcak order 2025-02-14 19:56:39 +03:00
Disty0 f94196bcd1 Rename ROCm Flash atten hijack to CK Flash atten and enable AOTriton memory and flash atten by default 2025-02-13 22:01:06 +03:00
Vladimir Mandic 49712ab9e7 update requirements
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-02-13 12:03:08 -05:00
Vladimir Mandic e28b8cd920 add torch gc debug
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-01-31 13:12:59 -05:00
Disty0 d2fee97e24 Update changelog 2025-01-31 20:27:22 +03:00
Disty0 039746914f Add check for missing cuda and ipex params 2025-01-31 19:12:56 +03:00
Vladimir Mandic 3dcb70e8a2 device init logging
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-01-31 09:31:12 -05:00
Vladimir Mandic 1697fb1508 add tunable ops path
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-01-30 13:35:09 -05:00
Vladimir Mandic 0ea7840608 add tunable ops
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-01-30 13:08:49 -05:00
Disty0 a770b1c888 More correct Dynamic Atten SDPA implementation and deprecate IPEX Diffusers attention 2025-01-25 21:33:42 +03:00
Disty0 af35296a68 IPEX 4GB alloc detection and log driver version 2025-01-22 18:15:25 +03:00
Vladimir Mandic bb97e695da log cleanup
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-12-31 09:15:36 -05:00
Vladimir Mandic 910f5d0a73 lora direct on-demand apply/unapply
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-12-29 12:38:19 -05:00
Vladimir Mandic 20f2554cec add sd35-ipadapter and more balanced offload optimizations
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-12-20 10:22:42 -05:00
Vladimir Mandic e9f951b2c5 offload logging
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-12-11 14:20:01 -05:00
Vladimir Mandic 9a588d9c91 update balanced offload
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-12-11 12:06:03 -05:00
Vladimir Mandic 023b13b6cb balanced offload improvements
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-12-01 15:34:25 -05:00
Vladimir Mandic b7aff134a2 add low/high threshold to balanced offload
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-11-30 19:03:51 -05:00
Vladimir Mandic b74166f9cb detailer add augment setting
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-11-29 07:18:07 -05:00
Vladimir Mandic dbb9ba0890 cuda memory limits
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-10-26 07:51:40 -04:00
Seunghoon Lee a76893bd72
add cdna check 2024-10-26 14:09:50 +09:00
Seunghoon Lee 81bd236cc3
zluda&rocm bf16 test 2024-10-26 13:59:21 +09:00
Disty0 d459acfcca Cleanup 2024-10-24 20:08:49 +03:00
Disty0 3b916d5e48 Zluda guess the GPU arch with the device name 2024-10-24 18:53:17 +03:00
Disty0 801ebdd080 Treat Zluda as a different backend and auto disable BF16 for Zluda and ROCm on RDNA1-2 2024-10-24 15:06:39 +03:00
Vladimir Mandic 0587f0be0c gc logging
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-10-23 07:58:09 -04:00
Vladimir Mandic 64f363283f messages,stats,save
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-10-20 09:26:07 -04:00
Disty0 6c11002420 PyTorch 2.5 XPU support 2024-10-17 23:11:52 +03:00
Disty0 b14e8f9a5f Don't assume Cuda on devices.same_device() 2024-10-14 17:23:51 +03:00
Vladimir Mandic c2ab0b11c3 check te device
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-10-14 09:29:33 -04:00
Disty0 9c05124c33 Add devices.has_xpu() 2024-10-13 15:13:38 +03:00
Disty0 011d9c3348 Move device backed initialization to shared.py 2024-10-13 14:56:28 +03:00
Disty0 84f8ab4076 Fix IPEX 2024-10-13 14:24:55 +03:00
Vladimir Mandic 0c54c235cb add sageattention
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-10-12 15:43:34 -04:00
Disty0 2e2cb43406 Make SDPA hijacks chainable and add Sage Attention 2024-10-12 21:19:38 +03:00
Vladimir Mandic ea0dfebe2d better handle any quant lib requirements
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-10-12 13:36:16 -04:00
Vladimir Mandic 3bbcc33181 add detailer
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-10-07 09:32:56 -04:00
Vladimir Mandic c21a10b7c9 add bf16 override for directml
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-09-29 21:34:59 -04:00
Vladimir Mandic b31d02ba1d cleanup
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-09-29 20:28:01 -04:00
Vladimir Mandic 47755dce6b refactor devices
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-09-29 20:17:03 -04:00
Vladimir Mandic fe94edf781 set default cuda dtype to auto
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-09-29 14:16:14 -04:00
Vladimir Mandic 174add0c3a restore dtype after upcast complete
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2024-09-29 13:22:08 -04:00
Vladimir Mandic 92f2a2902f improve profiling 2024-09-23 11:07:24 -04:00
Vladimir Mandic fe93ad6929 refactor xyzgrid 2024-09-13 19:22:02 -04:00
Disty0 586e5384b5 Update IPEX to 2.3 on Linux 2024-09-10 19:13:42 +03:00
Vladimir Mandic b4df9a4de1 jumbo update with flux.1 refactor, see changelog for details 2024-09-01 22:56:15 -04:00
Vladimir Mandic 5ed58ac7cc end-to-end update flux, see changelog and wiki 2024-08-28 08:04:24 -04:00
Disty0 a3f26c9df0 Convert Dynamic Attention SDP to a global SDP option 2024-08-18 01:44:27 +03:00
Disty0 f2769c0449 ROCm flash atten fall back to sdpa with fp32 inputs 2024-07-23 01:13:55 +03:00
xedis 809444eb5c
fix typo 2024-07-03 23:11:08 -07:00
Vladimir Mandic 16ab1a0af7 lint updates 2024-06-26 08:58:22 -04:00
Vladimir Mandic b036c2fc3b improve gc threshold 2024-06-21 12:57:17 -04:00
Disty0 092a326c09 Add torch_gc to state.nextjob, vae and upscale 2024-06-20 14:47:30 +03:00
Vladimir Mandic a1f53add94 fix typos 2024-06-16 17:00:35 -04:00
Vladimir Mandic 6d6f1de295 additional python 3.12 compatibility 2024-06-08 14:14:48 -04:00
Vladimir Mandic db9718eee6 add torch full deterministic mode 2024-06-07 09:26:51 -04:00
Vladimir Mandic d63f35e298 add cudaMallocAsync 2024-05-16 18:02:55 -04:00
Vladimir Mandic 40d5fdfdfd maybe fix slerp 2024-04-03 17:51:06 -04:00
Vladimir Mandic 834cb1b665 run fp16/bf16 test only once 2024-04-03 10:26:29 -04:00
Vladimir Mandic 9873178897 add extra_network_reference setting, refactor geninfo parser 2024-03-11 11:15:51 -04:00
Disty0 06149c4a41 ROCm add Flash Attention support 2024-03-10 00:11:50 +03:00
Vladimir Mandic ee7517dfb8 expose sdp options 2024-02-19 08:29:24 -05:00
Vladimir Mandic 1b3028b667 minor update 2024-02-15 09:13:34 -05:00
Vladimir Mandic d5a4f43f43 post release jumbo update 2024-02-08 12:10:32 -05:00
Disty0 0f829b2d04 Make OpenVINO compatible with IPEX venv 2024-01-08 03:00:18 +03:00
Disty0 8f70b7d08c Add DISABLE_VENV_LIBS env variable 2024-01-08 02:04:55 +03:00
Vladimir Mandic 17b30a320e enable batched taesd 2024-01-03 10:38:30 -05:00
Disty0 068f0a7d71 Return CPU device with OpenVINO on MAC 2024-01-02 20:20:36 +03:00
Vladimir Mandic 70bfe4ced8 enable gc on ram threshold 2023-12-31 08:15:22 -05:00
Vladimir Mandic 439542d3df redesign profiler 2023-12-03 11:27:24 -05:00
Disty0 bd141bbfeb IPEX decrease Torch GC Threshold to 80 2023-11-21 16:47:35 +03:00
Seunghoon Lee 36bef98cd5
Show device information log for DirectML. 2023-10-17 12:24:27 +09:00
Vladimir Mandic 2ec797472b add hypertile 2023-10-06 16:10:56 -04:00
Disty0 dc31dcbc1c Cleanup 2023-09-29 18:29:44 +03:00
Disty0 72a33d5247 Update device logging 2023-09-29 18:28:04 +03:00
Disty0 6184a8cb5c IPEX and DML fix Cuda error 2023-09-28 20:51:11 +03:00
Disty0 21d53b6ac8 Cleanup 2023-09-28 19:53:52 +03:00
Disty0 7a3c1da954 Add OpenVINO device logging 2023-09-28 19:33:14 +03:00
Disty0 5edf481c8d Add Torch GC threshold slider 2023-09-28 14:38:22 +03:00
Vladimir Mandic 0afcfe6097 logger early init 2023-09-23 23:44:34 -04:00
Disty0 550b7056ac IPEX fix SDPA and reduce torch_gc force to %90 2023-09-18 15:36:14 +03:00
Vladimir Mandic 484dae8dbd upgrade diffusers 2023-09-14 09:38:17 -04:00
Vladimir Mandic 76c444fbc8 cleanup 2023-09-13 11:48:13 -04:00
Vladimir Mandic f8fcb6f853 fix original hires non-latent 2023-09-10 18:30:20 -04:00
Vladimir Mandic 250d1bf2fb update hints 2023-09-10 13:05:31 -04:00
Disty0 34ee67477e Fix BF16 and FP32 logging 2023-09-08 23:49:49 +03:00
Vladimir Mandic 29d88cf557 cleanup logging 2023-09-08 13:29:33 -04:00
Vladimir Mandic f36c1eb476 jumbo patch 2023-09-08 13:01:20 -04:00
Vladimir Mandic 8fd96d0f30 catch directml and ipex initialization errors 2023-09-07 07:27:54 -04:00
Vladimir Mandic df65df3f36 minor fixes 2023-08-30 09:45:47 -04:00
Vladimir Mandic 48c0ce9b2b fix model lookups 2023-08-27 08:01:29 +00:00
Vladimir Mandic 6a4d4ea5b7 update logging and model hashinh 2023-08-22 18:28:09 +00:00
Disty0 f9718f068c Seperate OpenVINO from IPEX 2023-08-19 17:52:15 +03:00
Disty0 209f9a19c6 IPEX fixes 2023-08-16 18:56:50 +03:00
Vladimir Mandic 88fff06c9e downgrade warn to info 2023-08-13 10:58:02 +00:00
Vladimir Mandic 4234555566 update 2023-08-05 14:37:09 +00:00