SyCL intel build broken? #542
Comments
|
Similar issue on Arch Linux, also with intel-oneapi-basekit 2025.0.1. GPU is Battlemage B580 ./build/bin/sd --diffusion-model /mnt/hdd1/sd-models/flux1-dev-q4_k.gguf --vae /mnt/hdd1/sd-models/ae.safetensors --clip_l /mnt/hdd1/sd-models/clip_l.safetensors --t5xxl /mnt/hdd1/sd-models/t5xxl_fp16.safetensors -p "a lovely cat holding a sign says 'Battlemage B580'" --cfg-scale 1.0 --sampling-method euler_a -v
Option:
n_threads: 8
mode: txt2img
model_path:
wtype: unspecified
clip_l_path: /mnt/hdd1/sd-models/clip_l.safetensors
clip_g_path:
t5xxl_path: /mnt/hdd1/sd-models/t5xxl_fp16.safetensors
diffusion_model_path: /mnt/hdd1/sd-models/flux1-dev-q4_k.gguf
vae_path: /mnt/hdd1/sd-models/ae.safetensors
taesd_path:
esrgan_path:
controlnet_path:
embeddings_path:
stacked_id_embeddings_path:
input_id_images_path:
style ratio: 20.00
normalize input image : false
output_path: output.png
init_img:
mask_img:
control_image:
clip on cpu: false
controlnet cpu: false
vae decoder on cpu:false
diffusion flash attention:false
strength(control): 0.90
prompt: a lovely cat holding a sign says 'Battlemage B580'
negative_prompt:
min_cfg: 1.00
cfg_scale: 1.00
slg_scale: 0.00
guidance: 3.50
clip_skip: -1
width: 512
height: 512
sample_method: euler_a
schedule: default
sample_steps: 20
strength(img2img): 0.75
rng: cuda
seed: 42
batch_count: 1
vae_tiling: false
upscale_repeats: 1
System Info:
SSE3 = 1
AVX = 1
AVX2 = 1
AVX512 = 0
AVX512_VBMI = 0
AVX512_VNNI = 0
FMA = 1
NEON = 0
ARM_FMA = 0
F16C = 1
FP16_VA = 0
WASM_SIMD = 0
VSX = 0
[DEBUG] stable-diffusion.cpp:181 - Using SYCL backend
[SYCL] call ggml_check_sycl
ggml_check_sycl: GGML_SYCL_DEBUG: 0
ggml_check_sycl: GGML_SYCL_F16: no
found 1 SYCL devices:
| | | | |Max | |Max |Global | |
| | | | |compute|Max work|sub |mem | |
|ID| Device Type| Name|Version|units |group |group|size | Driver version|
|--|-------------------|---------------------------------------|-------|-------|--------|-----|-------|---------------------|
| 0| [level_zero:gpu:0]| Intel Graphics [0xe20b]| 20.1| 160| 1024| 32| 12168M| 1.6.31907|
ggml_sycl_init: GGML_SYCL_FORCE_MMQ: no
ggml_sycl_init: SYCL_USE_XMX: yes
ggml_sycl_init: found 1 SYCL devices:
[INFO ] stable-diffusion.cpp:202 - loading clip_l from '/mnt/hdd1/sd-models/clip_l.safetensors'
[INFO ] model.cpp:888 - load /mnt/hdd1/sd-models/clip_l.safetensors using safetensors format
[DEBUG] model.cpp:959 - init from '/mnt/hdd1/sd-models/clip_l.safetensors'
[INFO ] stable-diffusion.cpp:216 - loading t5xxl from '/mnt/hdd1/sd-models/t5xxl_fp16.safetensors'
[INFO ] model.cpp:888 - load /mnt/hdd1/sd-models/t5xxl_fp16.safetensors using safetensors format
[DEBUG] model.cpp:959 - init from '/mnt/hdd1/sd-models/t5xxl_fp16.safetensors'
[INFO ] stable-diffusion.cpp:223 - loading diffusion model from '/mnt/hdd1/sd-models/flux1-dev-q4_k.gguf'
[INFO ] model.cpp:885 - load /mnt/hdd1/sd-models/flux1-dev-q4_k.gguf using gguf format
[DEBUG] model.cpp:902 - init from '/mnt/hdd1/sd-models/flux1-dev-q4_k.gguf'
[INFO ] stable-diffusion.cpp:230 - loading vae from '/mnt/hdd1/sd-models/ae.safetensors'
[INFO ] model.cpp:888 - load /mnt/hdd1/sd-models/ae.safetensors using safetensors format
[DEBUG] model.cpp:959 - init from '/mnt/hdd1/sd-models/ae.safetensors'
[INFO ] stable-diffusion.cpp:242 - Version: Flux
[INFO ] stable-diffusion.cpp:275 - Weight type: f16
[INFO ] stable-diffusion.cpp:276 - Conditioner weight type: f16
[INFO ] stable-diffusion.cpp:277 - Diffusion model weight type: q4_K
[INFO ] stable-diffusion.cpp:278 - VAE weight type: f32
[DEBUG] stable-diffusion.cpp:280 - ggml tensor size = 400 bytes
[INFO ] stable-diffusion.cpp:319 - set clip_on_cpu to true
[INFO ] stable-diffusion.cpp:322 - CLIP: Using CPU backend
[DEBUG] clip.hpp:171 - vocab size: 49408
[DEBUG] clip.hpp:182 - trigger word img already in vocab
[INFO ] flux.hpp:889 - Flux blocks: 19 double, 38 single
[DEBUG] ggml_extend.hpp:1111 - clip params backend buffer size = 235.06 MB(RAM) (196 tensors)
[DEBUG] ggml_extend.hpp:1111 - t5 params backend buffer size = 9083.77 MB(RAM) (219 tensors)
[DEBUG] ggml_extend.hpp:1111 - flux params backend buffer size = 6604.64 MB(VRAM) (780 tensors)
[DEBUG] ggml_extend.hpp:1111 - vae params backend buffer size = 94.57 MB(VRAM) (138 tensors)
[DEBUG] stable-diffusion.cpp:417 - loading weights
[DEBUG] model.cpp:1698 - loading tensors from /mnt/hdd1/sd-models/clip_l.safetensors
|======> | 196/1440 - 0.00it/s[DEBUG] model.cpp:1698 - loading tensors from /mnt/hdd1/sd-models/t5xxl_fp16.safetensors
|==============> | 413/1440 - 0.00it/s[INFO ] model.cpp:1868 - unknown tensor 'text_encoders.t5xxl.transformer.encoder.embed_tokens.weight | f16 | 2 [4096, 32128, 1, 1, 1]' in model file
|==============> | 416/1440 - 9.43it/s[DEBUG] model.cpp:1698 - loading tensors from /mnt/hdd1/sd-models/flux1-dev-q4_k.gguf
|=========================================> | 1196/1440 - 50.00it/s[DEBUG] model.cpp:1698 - loading tensors from /mnt/hdd1/sd-models/ae.safetensors
|==============================================> | 1334/1440 - 200.00it/s[INFO ] stable-diffusion.cpp:516 - total params memory size = 16018.05MB (VRAM 6699.22MB, RAM 9318.83MB): clip 9318.83MB(RAM), unet 6604.64MB(VRAM), vae 94.57MB(VRAM), controlnet 0.00MB(VRAM), pmid 0.00MB(RAM)
[INFO ] stable-diffusion.cpp:520 - loading model from '' completed, taking 6.22s
[INFO ] stable-diffusion.cpp:537 - running in Flux FLOW mode
[DEBUG] stable-diffusion.cpp:594 - finished loaded file
[DEBUG] stable-diffusion.cpp:1535 - txt2img 512x512
[DEBUG] stable-diffusion.cpp:1230 - prompt after extract and remove lora: "a lovely cat holding a sign says 'Battlemage B580'"
[INFO ] stable-diffusion.cpp:682 - Attempting to apply 0 LoRAs
[INFO ] stable-diffusion.cpp:1235 - apply_loras completed, taking 0.00s
[DEBUG] conditioner.hpp:1027 - parse 'a lovely cat holding a sign says 'Battlemage B580'' to [['a lovely cat holding a sign says 'Battlemage B580'', 1], ]
[DEBUG] clip.hpp:311 - token length: 77
[DEBUG] t5.hpp:397 - token length: 256
[DEBUG] clip.hpp:736 - Missing text_projection matrix, assuming identity...
[DEBUG] ggml_extend.hpp:1062 - clip compute buffer size: 1.40 MB(RAM)
[DEBUG] clip.hpp:736 - Missing text_projection matrix, assuming identity...
[DEBUG] ggml_extend.hpp:1062 - t5 compute buffer size: 68.25 MB(RAM)
[DEBUG] conditioner.hpp:1142 - computing condition graph completed, taking 5611 ms
[INFO ] stable-diffusion.cpp:1368 - get_learned_condition completed, taking 5613 ms
[INFO ] stable-diffusion.cpp:1391 - sampling using Euler A method
[INFO ] stable-diffusion.cpp:1428 - generating image: 1/1 - seed 42
[DEBUG] stable-diffusion.cpp:798 - Sample
[DEBUG] ggml_extend.hpp:1062 - flux compute buffer size: 398.50 MB(VRAM)
No kernel named _ZTSZN14bin_bcast_syclIXadL_ZL6op_addffEEEclIfffEEvR25ggml_backend_sycl_contextPK11ggml_tensorS6_PS4_PKT_PKT0_PT1_PN4sycl3_V15queueEEUlNSH_7nd_itemILi3EEEE0_ was foundException caught at file:/mnt/hdd1/ipex/stable-diffusion.cpp/ggml/src/ggml-sycl/common.cpp, line:102 |
|
only version currently working for me is 0.2.1 from stable-diffusion-cpp-python |
Thanks for the heads-up! This one actually works :D |
|
this looks like a regression somewhere, 1c168d9 works |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Notes:
The text was updated successfully, but these errors were encountered: