Linux Mint Compile with TORCH_USE_CUDA_DSA to enable device-side assertions. Krita Stable Diffusion #2113
Description
I'm having some problems trying to make Krita SD work in Linux Mint, I'm a bit new in Linux world, so if you need more information, just ask it ...
Server Log
2025-10-22 02:45:39,736 INFO got prompt 2025-10-22 02:45:39,756 INFO 0 models unloaded. 2025-10-22 02:45:39,775 INFO loaded partially 2198.1744462013244 2198.174446105957 0 2025-10-22 02:45:39,775 INFO !!! Exception during processing !!! CUDA error: no kernel image is available for execution on the device 2025-10-22 02:45:39,776 INFO CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. 2025-10-22 02:45:39,776 INFO For debugging consider passing CUDA_LAUNCH_BLOCKING=1 2025-10-22 02:45:39,776 INFO Compile withTORCH_USE_CUDA_DSAto enable device-side assertions. 2025-10-22 02:45:39,776 INFO 2025-10-22 02:45:39,776 INFO Traceback (most recent call last): 2025-10-22 02:45:39,776 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/execution.py", line 496, in execute 2025-10-22 02:45:39,776 INFO output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs) 2025-10-22 02:45:39,776 INFO ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2025-10-22 02:45:39,776 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/execution.py", line 315, in get_output_data 2025-10-22 02:45:39,776 INFO return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs) 2025-10-22 02:45:39,777 INFO ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2025-10-22 02:45:39,777 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/execution.py", line 289, in _async_map_node_over_list 2025-10-22 02:45:39,777 INFO await process_inputs(input_dict, i) 2025-10-22 02:45:39,777 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/execution.py", line 277, in process_inputs 2025-10-22 02:45:39,777 INFO result = f(**inputs) 2025-10-22 02:45:39,777 INFO ^^^^^^^^^^^ 2025-10-22 02:45:39,777 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/comfy_extras/nodes_custom_sampler.py", line 835, in sample 2025-10-22 02:45:39,777 INFO samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed) 2025-10-22 02:45:39,777 INFO ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2025-10-22 02:45:39,777 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/comfy/samplers.py", line 1036, in sample 2025-10-22 02:45:39,778 INFO output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed) 2025-10-22 02:45:39,778 INFO ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2025-10-22 02:45:39,778 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/comfy/patcher_extension.py", line 112, in execute 2025-10-22 02:45:39,778 INFO return self.original(*args, **kwargs) 2025-10-22 02:45:39,778 INFO ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2025-10-22 02:45:39,778 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/comfy/samplers.py", line 1004, in outer_sample 2025-10-22 02:45:39,778 INFO output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed) 2025-10-22 02:45:39,778 INFO ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2025-10-22 02:45:39,778 INFO File "/home/bruno/.var/app/org.kde.krita/data/krita/ai_diffusion/server/ComfyUI/comfy/samplers.py", line 973, in inner_sample 2025-10-22 02:45:39,778 INFO if latent_image is not None and torch.count_nonzero(latent_image) > 0: #Don't shift the empty latent image. 2025-10-22 02:45:39,779 INFO ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 2025-10-22 02:45:39,779 INFO torch.AcceleratorError: CUDA error: no kernel image is available for execution on the device 2025-10-22 02:45:39,779 INFO CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. 2025-10-22 02:45:39,779 INFO For debugging consider passing CUDA_LAUNCH_BLOCKING=1 2025-10-22 02:45:39,779 INFO Compile withTORCH_USE_CUDA_DSA` to enable device-side assertions.
2025-10-22 02:45:39,779 INFO
2025-10-22 02:45:39,779 INFO
2025-10-22 02:45:39,779 INFO Prompt executed in 0.04 seconds
`
My GPU is a GTX 1050 Mobile and I'm in a Asus TFU Notebook and I'm using the local option (and NVIDIA CUDA)
this is all what I tried and nothing worked :
121 pip3 install numpy --pre torch torchvision torchaudio --force-reinstall --index-url https://download.pytorch.org/whl/nightly/cu117 122 pip3 install numpy --pre torch torchvision torchaudio --force-reinstall --index-url https://download.pytorch.org/whl/nightly/cu117 --break-system-packages 123 pip3 install numpy --pre torch torchvision torchaudio --force-reinstall --break-system-packages 124 pip uninstall torch 125 pip uninstall torch --break-system-packages~ 126 pip uninstall torch --break-system-packages && pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126 127 pip uninstall torch --break-system-packages && pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126 -y 128 pip uninstall torch --break-system-packages && pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126 129 pip uninstall torch --break-system-packages && pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126 --break-system-packages 130 history