PhotoMaker crash with quantized model

[rmatif]

When running PhotoMaker (I have just tested v1), it crashes when using a quantized model, while the fp16 models work just fine

@rmatif ➜ /workspaces/stable-diffusion.cpp (master) $ ./build/bin/sd -m models/sd_xl_turbo_1.0_fp16.safetensors  --taesd models/diffusion_pytorch_model.safetensors --stacked-id-embd-dir models/photomaker-v1.safetensors --input-id-images-dir photomaker -p "a girl img, retro futurism, retro game art style but extremely beautiful, intricate details, masterpiece, best quality, space-themed, cosmic, celestial, stars, galaxies, nebulas, planets, science fiction, highly detailed" --steps 1 --cfg-scale 1 --style-ratio 10 -o output.png -v -H 512 -W 512 --type q8_0
Option: 
    n_threads:         4
    mode:              txt2img
    model_path:        models/sd_xl_turbo_1.0_fp16.safetensors
    wtype:             q8_0
    clip_l_path:       
    clip_g_path:       
    t5xxl_path:        
    diffusion_model_path:   
    vae_path:          
    taesd_path:        models/diffusion_pytorch_model.safetensors
    esrgan_path:       
    controlnet_path:   
    embeddings_path:   
    stacked_id_embeddings_path:   models/photomaker-v1.safetensors
    input_id_images_path:   photomaker
    style ratio:       10.00
    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 girl img, retro futurism, retro game art style but extremely beautiful, intricate details, masterpiece, best quality, space-themed, cosmic, celestial, stars, galaxies, nebulas, planets, science fiction, highly detailed
    negative_prompt:   
    min_cfg:           1.00
    cfg_scale:         1.00
    slg_scale:         0.00
    guidance:          3.50
    eta:               0.00
    clip_skip:         -1
    width:             512
    height:            512
    sample_method:     euler_a
    schedule:          default
    sample_steps:      1
    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:188  - Using CPU backend
[INFO ] stable-diffusion.cpp:197  - loading model from 'models/sd_xl_turbo_1.0_fp16.safetensors'
[INFO ] model.cpp:908  - load models/sd_xl_turbo_1.0_fp16.safetensors using safetensors format
[DEBUG] model.cpp:979  - init from 'models/sd_xl_turbo_1.0_fp16.safetensors'
[INFO ] stable-diffusion.cpp:244  - Version: SDXL 
[INFO ] stable-diffusion.cpp:277  - Weight type:                 q8_0
[INFO ] stable-diffusion.cpp:278  - Conditioner weight type:     q8_0
[INFO ] stable-diffusion.cpp:279  - Diffusion model weight type: q8_0
[INFO ] stable-diffusion.cpp:280  - VAE weight type:             f32
[DEBUG] stable-diffusion.cpp:282  - ggml tensor size = 400 bytes
[DEBUG] clip.hpp:171  - vocab size: 49408
[DEBUG] clip.hpp:182  -  trigger word img already in vocab
[DEBUG] ggml_extend.hpp:1174 - clip params backend buffer size =  231.50 MB(RAM) (196 tensors)
[DEBUG] ggml_extend.hpp:1174 - clip params backend buffer size =  887.42 MB(RAM) (517 tensors)
[DEBUG] ggml_extend.hpp:1174 - unet params backend buffer size =  2931.68 MB(RAM) (1680 tensors)
[INFO ] model.cpp:908  - load models/photomaker-v1.safetensors using safetensors format
[DEBUG] model.cpp:979  - init from 'models/photomaker-v1.safetensors'
[INFO ] lora.hpp:117  - loading LoRA from 'models/photomaker-v1.safetensors'
[DEBUG] model.cpp:1727 - loading tensors from models/photomaker-v1.safetensors
[DEBUG] ggml_extend.hpp:1174 - lora params backend buffer size =  354.38 MB(RAM) (1120 tensors)
[DEBUG] model.cpp:1727 - loading tensors from models/photomaker-v1.safetensors
  |=====================================>            | 1134/1527 - 1000.00it/s[DEBUG] lora.hpp:160  - lora type: ".lora_down"/".lora_up"
[DEBUG] lora.hpp:162  - finished loaded lora
[INFO ] stable-diffusion.cpp:392  - loading stacked ID embedding (PHOTOMAKER) model file from 'models/photomaker-v1.safetensors'
[INFO ] model.cpp:908  - load models/photomaker-v1.safetensors using safetensors format
[DEBUG] model.cpp:979  - init from 'models/photomaker-v1.safetensors'
[DEBUG] ggml_extend.hpp:1174 - pmid params backend buffer size =  1243.48 MB(RAM) (407 tensors)
[DEBUG] stable-diffusion.cpp:419  - loading weights
[DEBUG] model.cpp:1727 - loading tensors from models/sd_xl_turbo_1.0_fp16.safetensors
  |===============================>                  | 2641/4168 - 125.00it/s[DEBUG] model.cpp:1727 - loading tensors from models/photomaker-v1.safetensors
  |==================================================| 4168/4168 - 90.91it/s
[INFO ] tae.hpp:214  - loading taesd from 'models/diffusion_pytorch_model.safetensors', decode_only = true
[DEBUG] ggml_extend.hpp:1174 - taesd params backend buffer size =   2.34 MB(RAM) (67 tensors)
[INFO ] model.cpp:908  - load models/diffusion_pytorch_model.safetensors using safetensors format
[DEBUG] model.cpp:979  - init from 'models/diffusion_pytorch_model.safetensors'
[DEBUG] model.cpp:1727 - loading tensors from models/diffusion_pytorch_model.safetensors
  |=========================>                        | 67/134 - 0.00it/s[INFO ] tae.hpp:236  - taesd model loaded
[INFO ] stable-diffusion.cpp:503  - total params memory size = 5296.42MB (VRAM 2.34MB, RAM 5294.09MB): clip 1118.92MB(RAM), unet 2931.68MB(RAM), vae 2.34MB(VRAM), controlnet 0.00MB(VRAM), pmid 1243.48MB(RAM)
[INFO ] stable-diffusion.cpp:522  - loading model from 'models/sd_xl_turbo_1.0_fp16.safetensors' completed, taking 29.53s
[INFO ] stable-diffusion.cpp:556  - running in eps-prediction mode
[DEBUG] stable-diffusion.cpp:600  - finished loaded file
[DEBUG] stable-diffusion.cpp:1548 - txt2img 512x512
[DEBUG] stable-diffusion.cpp:1241 - prompt after extract and remove lora: "a girl img, retro futurism, retro game art style but extremely beautiful, intricate details, masterpiece, best quality, space-themed, cosmic, celestial, stars, galaxies, nebulas, planets, science fiction, highly detailed"
[INFO ] stable-diffusion.cpp:690  - Attempting to apply 0 LoRAs
[INFO ] stable-diffusion.cpp:1246 - apply_loras completed, taking 0.00s
/workspaces/stable-diffusion.cpp/ggml/src/ggml.c:5764: GGML_ASSERT(cgraph->n_nodes < cgraph->size) failed
Could not attach to process.  If your uid matches the uid of the target
process, check the setting of /proc/sys/kernel/yama/ptrace_scope, or try
again as the root user.  For more details, see /etc/sysctl.d/10-ptrace.conf
ptrace: Operation not permitted.
No stack.
The program is not being run.
Aborted (core dumped)

The same command without --type q8_0 works just fine

[wbruna]

Looks like the same assertion failure reported as #684 and #688 . Should be fixed by #658 .