Comfy-Org/comfy-aimdo
1
0
Fork 0
mirror of https://github.com/Comfy-Org/comfy-aimdo.git synced 2026-11-04 23:37:21 +00:00
Code Issues 43 Projects Packages Wiki Activity

[GH-ISSUE #45] OSError: exception: access violation reading 0x00000000000000E0 #15

New issue
Closed
opened 2026-10-26 22:23:36 +00:00 by bsakaguchi · 2 comments
bsakaguchi commented 2026-10-26 22:23:36 +00:00
Owner
Copy link

Originally created by @alexsarmiento on GitHub (May 4, 2026).
Original GitHub issue: https://github.com/Comfy-Org/comfy-aimdo/issues/45

Using a Klein-9b model comfy-aimdo fails after one or two runs. AMD RX 7900 XTX on Windows

Total VRAM 24560 MB, total RAM 93573 MB
pytorch version: 2.10.0+rocm7.13.0a20260503
Set: torch.backends.cudnn.enabled = False for better AMD performance.
AMD arch: gfx1100
ROCm version: (7, 13)
Set vram state to: NORMAL_VRAM
Disabling smart memory management
Device: cuda:0 AMD Radeon RX 7900 XTX : native
Using async weight offloading with 2 streams
Using sage attention
aimdo: src-win/cuda-detour.c:38:INFO:aimdo_setup_hooks: installing 6 hooks
aimdo: src-win/cuda-detour.c:28:DEBUG:install_hook_entries: hooks successfully installed
aimdo: src-win/shmem-detect.c:80:INFO:comfy-aimdo WDDM adapter match: AMD Radeon RX 7900 XTX runtime_luid=00000000:000115f8 dxgi_luid=00000000:000115f8
aimdo: src/control.c:152:INFO:comfy-aimdo inited for GPU: AMD Radeon RX 7900 XTX (VRAM: 24560 MB)
DynamicVRAM support detected and enabled
Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr  8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)]
ComfyUI version: 0.20.1
comfy-aimdo version: 0.3.0
comfy-kitchen version: 0.2.8
ComfyUI frontend version: 1.42.15

got prompt
Requested to load AutoencoderKL
Model AutoencoderKL prepared for dynamic VRAM loading. 160MB Staged. 0 patches attached.
Requested to load Flux2
Model Flux2 prepared for dynamic VRAM loading. 9256MB Staged. 112 patches attached.
  0%|                                                                | 0/6 [00:00<?, ?it/s,   Model Initializing ...  ]Windows fatal exception: access violation

Stack (most recent call first):
  File "D:\ComfyUI\venv\Lib\site-packages\comfy_aimdo\vram_buffer.py", line 31 in __init__
  File "D:\ComfyUI\comfy\model_management.py", line 1219 in get_aimdo_cast_buffer
  File "D:\ComfyUI\comfy\ops.py", line 122 in get_cast_buffer
  File "D:\ComfyUI\comfy\ops.py", line 181 in <lambda>
  File "D:\ComfyUI\comfy\lora.py", line 480 in prefetch_prepared_value
  File "D:\ComfyUI\comfy\lora.py", line 486 in <genexpr>
  File "D:\ComfyUI\comfy\lora.py", line 486 in prefetch_prepared_value
  File "D:\ComfyUI\comfy\lora.py", line 484 in prefetch_prepared_value
  File "D:\ComfyUI\comfy\model_patcher.py", line 128 in prepare
  File "D:\ComfyUI\comfy\ops.py", line 181 in cast_modules_with_vbar
  File "D:\ComfyUI\comfy\ops.py", line 299 in cast_bias_weight
  File "D:\ComfyUI\comfy\ops.py", line 1149 in forward_comfy_cast_weights
  File "D:\ComfyUI\comfy\ops.py", line 1206 in forward
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787 in _call_impl
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776 in _wrapped_call_impl
  File "D:\ComfyUI\comfy\ldm\flux\layers.py", line 214 in forward
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787 in _call_impl
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776 in _wrapped_call_impl
  File "D:\ComfyUI\comfy\ldm\flux\model.py", line 243 in forward_orig
  File "D:\ComfyUI\comfy\ldm\flux\model.py", line 406 in _forward
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute
  File "D:\ComfyUI\comfy\ldm\flux\model.py", line 345 in forward
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787 in _call_impl
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776 in _wrapped_call_impl
  File "D:\ComfyUI\comfy\model_base.py", line 222 in _apply_model
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute
  File "D:\ComfyUI\comfy\model_base.py", line 178 in apply_model
  File "D:\ComfyUI\comfy\samplers.py", line 325 in _calc_cond_batch
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute
  File "D:\ComfyUI\comfy\samplers.py", line 213 in _calc_cond_batch_outer
  File "D:\ComfyUI\comfy\samplers.py", line 205 in calc_cond_batch
  File "D:\ComfyUI\comfy\samplers.py", line 380 in sampling_function
  File "D:\ComfyUI\comfy\samplers.py", line 964 in predict_noise
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute
  File "D:\ComfyUI\comfy\samplers.py", line 961 in outer_predict_noise
  File "D:\ComfyUI\comfy\samplers.py", line 954 in __call__
  File "D:\ComfyUI\comfy\samplers.py", line 400 in __call__
  File "D:\ComfyUI\comfy\k_diffusion\sampling.py", line 205 in sample_euler
  File "D:\ComfyUI\venv\Lib\site-packages\torch\utils\_contextlib.py", line 124 in decorate_context
  File "D:\ComfyUI\comfy\samplers.py", line 751 in sample
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute
  File "D:\ComfyUI\comfy\samplers.py", line 981 in inner_sample
  File "D:\ComfyUI\comfy\samplers.py", line 995 in outer_sample
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute
  File "D:\ComfyUI\comfy\samplers.py", line 1052 in sample
  File "D:\ComfyUI\comfy_extras\nodes_custom_sampler.py", line 963 in execute
  File "D:\ComfyUI\comfy_api\latest\_io.py", line 1826 in EXECUTE_NORMALIZED
  File "D:\ComfyUI\comfy_api\internal\__init__.py", line 149 in wrapped_func
  File "D:\ComfyUI\execution.py", line 297 in process_inputs
  File "D:\ComfyUI\execution.py", line 309 in _async_map_node_over_list
  File "D:\ComfyUI\execution.py", line 335 in get_output_data
  File "D:\ComfyUI\execution.py", line 535 in execute
  File "D:\ComfyUI\execution.py", line 772 in execute_async
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\events.py", line 88 in _run
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 1999 in _run_once
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 645 in run_forever
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\windows_events.py", line 322 in run_forever
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 678 in run_until_complete
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 118 in run
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 195 in run
  File "D:\ComfyUI\execution.py", line 713 in execute
  File "D:\ComfyUI\main.py", line 313 in prompt_worker
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\threading.py", line 1012 in run
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\threading.py", line 1075 in _bootstrap_inner
  File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\threading.py", line 1032 in _bootstrap
  0%|                                                                | 0/6 [00:00<?, ?it/s,   Model Initializing ...  ]
!!! Exception during processing !!! exception: access violation reading 0x00000000000000E0
Traceback (most recent call last):
  File "D:\ComfyUI\execution.py", line 535, in execute
    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, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\execution.py", line 335, in get_output_data
    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, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\execution.py", line 309, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "D:\ComfyUI\execution.py", line 297, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "D:\ComfyUI\comfy_api\internal\__init__.py", line 149, in wrapped_func
    return method(locked_class, **inputs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy_api\latest\_io.py", line 1826, in EXECUTE_NORMALIZED
    to_return = cls.execute(*args, **kwargs)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy_extras\nodes_custom_sampler.py", line 963, in execute
    samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 1052, in sample
    output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute
    return self.original(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 995, in outer_sample
    output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 981, in inner_sample
    samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute
    return self.original(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 751, in sample
    samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\torch\utils\_contextlib.py", line 124, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\k_diffusion\sampling.py", line 205, in sample_euler
    denoised = model(x, sigma_hat * s_in, **extra_args)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 400, in __call__
    out = self.inner_model(x, sigma, model_options=model_options, seed=seed)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 954, in __call__
    return self.outer_predict_noise(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 961, in outer_predict_noise
    ).execute(x, timestep, model_options, seed)
      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute
    return self.original(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 964, in predict_noise
    return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 380, in sampling_function
    out = calc_cond_batch(model, conds, x, timestep, model_options)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 205, in calc_cond_batch
    return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 213, in _calc_cond_batch_outer
    return executor.execute(model, conds, x_in, timestep, model_options)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute
    return self.original(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\samplers.py", line 325, in _calc_cond_batch
    output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\model_base.py", line 178, in apply_model
    return comfy.patcher_extension.WrapperExecutor.new_class_executor(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute
    return self.original(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\model_base.py", line 222, in _apply_model
    model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ldm\flux\model.py", line 345, in forward
    return comfy.patcher_extension.WrapperExecutor.new_class_executor(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute
    return self.original(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ldm\flux\model.py", line 406, in _forward
    out = self.forward_orig(img, img_ids, context, txt_ids, timestep, y, guidance, control, timestep_zero_index=timestep_zero_index, transformer_options=transformer_options, attn_mask=kwargs.get("attention_mask", None))
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ldm\flux\model.py", line 243, in forward_orig
    img, txt = block(img=img,
               ^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ldm\flux\layers.py", line 214, in forward
    txt_qkv = self.txt_attn.qkv(txt_modulated)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ops.py", line 1206, in forward
    output = self.forward_comfy_cast_weights(input, compute_dtype, want_requant=isinstance(input, QuantizedTensor))
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ops.py", line 1149, in forward_comfy_cast_weights
    weight, bias, offload_stream = cast_bias_weight(self, input, offloadable=True, compute_dtype=compute_dtype, want_requant=want_requant)
                                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ops.py", line 299, in cast_bias_weight
    offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ops.py", line 181, in cast_modules_with_vbar
    lowvram_fn.prepare(lambda size: get_cast_buffer(size), offload_stream)
  File "D:\ComfyUI\comfy\model_patcher.py", line 128, in prepare
    (patch[0], comfy.lora.prefetch_prepared_value(patch[1], allocate_buffer, stream), patch[2], patch[3], patch[4])
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\lora.py", line 484, in prefetch_prepared_value
    return type(value)(value.loaded_keys, prefetch_prepared_value(value.weights, allocate_buffer, stream))
                                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\lora.py", line 486, in prefetch_prepared_value
    return tuple(prefetch_prepared_value(item, allocate_buffer, stream) for item in value)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\lora.py", line 486, in <genexpr>
    return tuple(prefetch_prepared_value(item, allocate_buffer, stream) for item in value)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\lora.py", line 480, in prefetch_prepared_value
    dest = allocate_buffer(comfy.memory_management.vram_aligned_size(value))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ops.py", line 181, in <lambda>
    lowvram_fn.prepare(lambda size: get_cast_buffer(size), offload_stream)
                                    ^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\ops.py", line 122, in get_cast_buffer
    cast_buffer = comfy.model_management.get_aimdo_cast_buffer(offload_stream, device)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\comfy\model_management.py", line 1219, in get_aimdo_cast_buffer
    cast_buffer = comfy_aimdo.vram_buffer.VRAMBuffer(DEFAULT_AIMDO_CAST_BUFFER_RESERVATION_SIZE, device.index)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ComfyUI\venv\Lib\site-packages\comfy_aimdo\vram_buffer.py", line 31, in __init__
    self._ptr = lib.vrambuf_create(self.device, self.max_size)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: exception: access violation reading 0x00000000000000E0
Originally created by @alexsarmiento on GitHub (May 4, 2026). Original GitHub issue: https://github.com/Comfy-Org/comfy-aimdo/issues/45 Using a Klein-9b model comfy-aimdo fails after one or two runs. AMD RX 7900 XTX on Windows ``` Total VRAM 24560 MB, total RAM 93573 MB pytorch version: 2.10.0+rocm7.13.0a20260503 Set: torch.backends.cudnn.enabled = False for better AMD performance. AMD arch: gfx1100 ROCm version: (7, 13) Set vram state to: NORMAL_VRAM Disabling smart memory management Device: cuda:0 AMD Radeon RX 7900 XTX : native Using async weight offloading with 2 streams Using sage attention aimdo: src-win/cuda-detour.c:38:INFO:aimdo_setup_hooks: installing 6 hooks aimdo: src-win/cuda-detour.c:28:DEBUG:install_hook_entries: hooks successfully installed aimdo: src-win/shmem-detect.c:80:INFO:comfy-aimdo WDDM adapter match: AMD Radeon RX 7900 XTX runtime_luid=00000000:000115f8 dxgi_luid=00000000:000115f8 aimdo: src/control.c:152:INFO:comfy-aimdo inited for GPU: AMD Radeon RX 7900 XTX (VRAM: 24560 MB) DynamicVRAM support detected and enabled Python version: 3.12.10 (tags/v3.12.10:0cc8128, Apr 8 2025, 12:21:36) [MSC v.1943 64 bit (AMD64)] ComfyUI version: 0.20.1 comfy-aimdo version: 0.3.0 comfy-kitchen version: 0.2.8 ComfyUI frontend version: 1.42.15 ``` ``` got prompt Requested to load AutoencoderKL Model AutoencoderKL prepared for dynamic VRAM loading. 160MB Staged. 0 patches attached. Requested to load Flux2 Model Flux2 prepared for dynamic VRAM loading. 9256MB Staged. 112 patches attached. 0%| | 0/6 [00:00<?, ?it/s, Model Initializing ... ]Windows fatal exception: access violation Stack (most recent call first): File "D:\ComfyUI\venv\Lib\site-packages\comfy_aimdo\vram_buffer.py", line 31 in __init__ File "D:\ComfyUI\comfy\model_management.py", line 1219 in get_aimdo_cast_buffer File "D:\ComfyUI\comfy\ops.py", line 122 in get_cast_buffer File "D:\ComfyUI\comfy\ops.py", line 181 in <lambda> File "D:\ComfyUI\comfy\lora.py", line 480 in prefetch_prepared_value File "D:\ComfyUI\comfy\lora.py", line 486 in <genexpr> File "D:\ComfyUI\comfy\lora.py", line 486 in prefetch_prepared_value File "D:\ComfyUI\comfy\lora.py", line 484 in prefetch_prepared_value File "D:\ComfyUI\comfy\model_patcher.py", line 128 in prepare File "D:\ComfyUI\comfy\ops.py", line 181 in cast_modules_with_vbar File "D:\ComfyUI\comfy\ops.py", line 299 in cast_bias_weight File "D:\ComfyUI\comfy\ops.py", line 1149 in forward_comfy_cast_weights File "D:\ComfyUI\comfy\ops.py", line 1206 in forward File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787 in _call_impl File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776 in _wrapped_call_impl File "D:\ComfyUI\comfy\ldm\flux\layers.py", line 214 in forward File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787 in _call_impl File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776 in _wrapped_call_impl File "D:\ComfyUI\comfy\ldm\flux\model.py", line 243 in forward_orig File "D:\ComfyUI\comfy\ldm\flux\model.py", line 406 in _forward File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute File "D:\ComfyUI\comfy\ldm\flux\model.py", line 345 in forward File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787 in _call_impl File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776 in _wrapped_call_impl File "D:\ComfyUI\comfy\model_base.py", line 222 in _apply_model File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute File "D:\ComfyUI\comfy\model_base.py", line 178 in apply_model File "D:\ComfyUI\comfy\samplers.py", line 325 in _calc_cond_batch File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute File "D:\ComfyUI\comfy\samplers.py", line 213 in _calc_cond_batch_outer File "D:\ComfyUI\comfy\samplers.py", line 205 in calc_cond_batch File "D:\ComfyUI\comfy\samplers.py", line 380 in sampling_function File "D:\ComfyUI\comfy\samplers.py", line 964 in predict_noise File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute File "D:\ComfyUI\comfy\samplers.py", line 961 in outer_predict_noise File "D:\ComfyUI\comfy\samplers.py", line 954 in __call__ File "D:\ComfyUI\comfy\samplers.py", line 400 in __call__ File "D:\ComfyUI\comfy\k_diffusion\sampling.py", line 205 in sample_euler File "D:\ComfyUI\venv\Lib\site-packages\torch\utils\_contextlib.py", line 124 in decorate_context File "D:\ComfyUI\comfy\samplers.py", line 751 in sample File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute File "D:\ComfyUI\comfy\samplers.py", line 981 in inner_sample File "D:\ComfyUI\comfy\samplers.py", line 995 in outer_sample File "D:\ComfyUI\comfy\patcher_extension.py", line 112 in execute File "D:\ComfyUI\comfy\samplers.py", line 1052 in sample File "D:\ComfyUI\comfy_extras\nodes_custom_sampler.py", line 963 in execute File "D:\ComfyUI\comfy_api\latest\_io.py", line 1826 in EXECUTE_NORMALIZED File "D:\ComfyUI\comfy_api\internal\__init__.py", line 149 in wrapped_func File "D:\ComfyUI\execution.py", line 297 in process_inputs File "D:\ComfyUI\execution.py", line 309 in _async_map_node_over_list File "D:\ComfyUI\execution.py", line 335 in get_output_data File "D:\ComfyUI\execution.py", line 535 in execute File "D:\ComfyUI\execution.py", line 772 in execute_async File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\events.py", line 88 in _run File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 1999 in _run_once File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 645 in run_forever File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\windows_events.py", line 322 in run_forever File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 678 in run_until_complete File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 118 in run File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 195 in run File "D:\ComfyUI\execution.py", line 713 in execute File "D:\ComfyUI\main.py", line 313 in prompt_worker File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\threading.py", line 1012 in run File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\threading.py", line 1075 in _bootstrap_inner File "C:\Users\sarmi\AppData\Local\Programs\Python\Python312\Lib\threading.py", line 1032 in _bootstrap 0%| | 0/6 [00:00<?, ?it/s, Model Initializing ... ] !!! Exception during processing !!! exception: access violation reading 0x00000000000000E0 Traceback (most recent call last): File "D:\ComfyUI\execution.py", line 535, in execute 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, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\execution.py", line 335, in get_output_data 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, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\execution.py", line 309, in _async_map_node_over_list await process_inputs(input_dict, i) File "D:\ComfyUI\execution.py", line 297, in process_inputs result = f(**inputs) ^^^^^^^^^^^ File "D:\ComfyUI\comfy_api\internal\__init__.py", line 149, in wrapped_func return method(locked_class, **inputs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy_api\latest\_io.py", line 1826, in EXECUTE_NORMALIZED to_return = cls.execute(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy_extras\nodes_custom_sampler.py", line 963, in execute samples = guider.sample(noise.generate_noise(latent), latent_image, sampler, sigmas, denoise_mask=noise_mask, callback=callback, disable_pbar=disable_pbar, seed=noise.seed) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 1052, in sample output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 995, in outer_sample output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 981, in inner_sample samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 751, in sample samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\torch\utils\_contextlib.py", line 124, in decorate_context return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\k_diffusion\sampling.py", line 205, in sample_euler denoised = model(x, sigma_hat * s_in, **extra_args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 400, in __call__ out = self.inner_model(x, sigma, model_options=model_options, seed=seed) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 954, in __call__ return self.outer_predict_noise(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 961, in outer_predict_noise ).execute(x, timestep, model_options, seed) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 964, in predict_noise return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 380, in sampling_function out = calc_cond_batch(model, conds, x, timestep, model_options) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 205, in calc_cond_batch return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 213, in _calc_cond_batch_outer return executor.execute(model, conds, x_in, timestep, model_options) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\samplers.py", line 325, in _calc_cond_batch output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\model_base.py", line 178, in apply_model return comfy.patcher_extension.WrapperExecutor.new_class_executor( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\model_base.py", line 222, in _apply_model model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776, in _wrapped_call_impl return self._call_impl(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ldm\flux\model.py", line 345, in forward return comfy.patcher_extension.WrapperExecutor.new_class_executor( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ldm\flux\model.py", line 406, in _forward out = self.forward_orig(img, img_ids, context, txt_ids, timestep, y, guidance, control, timestep_zero_index=timestep_zero_index, transformer_options=transformer_options, attn_mask=kwargs.get("attention_mask", None)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ldm\flux\model.py", line 243, in forward_orig img, txt = block(img=img, ^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776, in _wrapped_call_impl return self._call_impl(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ldm\flux\layers.py", line 214, in forward txt_qkv = self.txt_attn.qkv(txt_modulated) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1776, in _wrapped_call_impl return self._call_impl(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\torch\nn\modules\module.py", line 1787, in _call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ops.py", line 1206, in forward output = self.forward_comfy_cast_weights(input, compute_dtype, want_requant=isinstance(input, QuantizedTensor)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ops.py", line 1149, in forward_comfy_cast_weights weight, bias, offload_stream = cast_bias_weight(self, input, offloadable=True, compute_dtype=compute_dtype, want_requant=want_requant) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ops.py", line 299, in cast_bias_weight offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ops.py", line 181, in cast_modules_with_vbar lowvram_fn.prepare(lambda size: get_cast_buffer(size), offload_stream) File "D:\ComfyUI\comfy\model_patcher.py", line 128, in prepare (patch[0], comfy.lora.prefetch_prepared_value(patch[1], allocate_buffer, stream), patch[2], patch[3], patch[4]) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\lora.py", line 484, in prefetch_prepared_value return type(value)(value.loaded_keys, prefetch_prepared_value(value.weights, allocate_buffer, stream)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\lora.py", line 486, in prefetch_prepared_value return tuple(prefetch_prepared_value(item, allocate_buffer, stream) for item in value) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\lora.py", line 486, in <genexpr> return tuple(prefetch_prepared_value(item, allocate_buffer, stream) for item in value) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\lora.py", line 480, in prefetch_prepared_value dest = allocate_buffer(comfy.memory_management.vram_aligned_size(value)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ops.py", line 181, in <lambda> lowvram_fn.prepare(lambda size: get_cast_buffer(size), offload_stream) ^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\ops.py", line 122, in get_cast_buffer cast_buffer = comfy.model_management.get_aimdo_cast_buffer(offload_stream, device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\model_management.py", line 1219, in get_aimdo_cast_buffer cast_buffer = comfy_aimdo.vram_buffer.VRAMBuffer(DEFAULT_AIMDO_CAST_BUFFER_RESERVATION_SIZE, device.index) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\venv\Lib\site-packages\comfy_aimdo\vram_buffer.py", line 31, in __init__ self._ptr = lib.vrambuf_create(self.device, self.max_size) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ OSError: exception: access violation reading 0x00000000000000E0 ```
bsakaguchi commented 2026-10-26 22:23:37 +00:00
Author
Owner
Copy link

@Apophis3158 commented on GitHub (May 12, 2026):

access violation reading 0x00000000000000E0 is a rocm internal issue, addressed in https://github.com/ROCm/rocm-systems/pull/6051.

After this fix I got:

0 models unloaded.
Model AutoencoderKL prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached.
aimdo: src/vrambuf.c:42:ERROR:vrambuf_create: 0 16777216k
!!! Exception during processing !!! VRAM reservation failed
Traceback (most recent call last):
  File "H:\ROCm\ComfyUI\execution.py", line 535, in execute
    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, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\execution.py", line 335, in get_output_data
    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, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\execution.py", line 309, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "H:\ROCm\ComfyUI\execution.py", line 297, in process_inputs
    result = f(**inputs)
  File "H:\ROCm\ComfyUI\nodes.py", line 316, in decode
    images = vae.decode(latent)
  File "H:\ROCm\ComfyUI\comfy\sd.py", line 1014, in decode
    model_management.raise_non_oom(e)
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
  File "H:\ROCm\ComfyUI\comfy\model_management.py", line 286, in raise_non_oom
    raise e
  File "H:\ROCm\ComfyUI\comfy\sd.py", line 1007, in decode
    out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True)
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\comfy\ldm\models\autoencoder.py", line 254, in decode
    dec = self.decoder(dec, **decoder_kwargs)
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl
    return forward_call(*args, **kwargs)
  File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 808, in forward
    h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs)
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl
    return forward_call(*args, **kwargs)
  File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 224, in forward
    h = conv_carry_causal_3d(h, self.conv2, conv_carry_in=conv_carry_in, conv_carry_out=conv_carry_out)
  File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 81, in conv_carry_causal_3d
    out = op(x)
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl
    return forward_call(*args, **kwargs)
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 516, in forward
    return self.forward_comfy_cast_weights(*args, **kwargs)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 508, in forward_comfy_cast_weights
    weight, bias, offload_stream = cast_bias_weight(self, input, offloadable=True)
                                   ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 302, in cast_bias_weight
    offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking)
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 163, in cast_modules_with_vbar
    xfer_dest = get_cast_buffer(dest_size)
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 122, in get_cast_buffer
    cast_buffer = comfy.model_management.get_aimdo_cast_buffer(offload_stream, device)
  File "H:\ROCm\ComfyUI\comfy\model_management.py", line 1221, in get_aimdo_cast_buffer
    cast_buffer = comfy_aimdo.vram_buffer.VRAMBuffer(DEFAULT_AIMDO_CAST_BUFFER_RESERVATION_SIZE, device.index)
  File "H:\ROCm\comfy-aimdo\comfy_aimdo\vram_buffer.py", line 34, in __init__
    raise RuntimeError("VRAM reservation failed")
RuntimeError: VRAM reservation failed

which looks similar to CUDA's https://github.com/Comfy-Org/comfy-aimdo/issues/49.

<!-- gh-comment-id:4435751398 --> @Apophis3158 commented on GitHub (May 12, 2026): `access violation reading 0x00000000000000E0` is a rocm internal issue, addressed in https://github.com/ROCm/rocm-systems/pull/6051. After this fix I got: ``` 0 models unloaded. Model AutoencoderKL prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached. aimdo: src/vrambuf.c:42:ERROR:vrambuf_create: 0 16777216k !!! Exception during processing !!! VRAM reservation failed Traceback (most recent call last): File "H:\ROCm\ComfyUI\execution.py", line 535, in execute 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, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ROCm\ComfyUI\execution.py", line 335, in get_output_data 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, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ROCm\ComfyUI\execution.py", line 309, in _async_map_node_over_list await process_inputs(input_dict, i) File "H:\ROCm\ComfyUI\execution.py", line 297, in process_inputs result = f(**inputs) File "H:\ROCm\ComfyUI\nodes.py", line 316, in decode images = vae.decode(latent) File "H:\ROCm\ComfyUI\comfy\sd.py", line 1014, in decode model_management.raise_non_oom(e) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^ File "H:\ROCm\ComfyUI\comfy\model_management.py", line 286, in raise_non_oom raise e File "H:\ROCm\ComfyUI\comfy\sd.py", line 1007, in decode out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ROCm\ComfyUI\comfy\ldm\models\autoencoder.py", line 254, in decode dec = self.decoder(dec, **decoder_kwargs) File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl return self._call_impl(*args, **kwargs) ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl return forward_call(*args, **kwargs) File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 808, in forward h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs) File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl return self._call_impl(*args, **kwargs) ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl return forward_call(*args, **kwargs) File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 224, in forward h = conv_carry_causal_3d(h, self.conv2, conv_carry_in=conv_carry_in, conv_carry_out=conv_carry_out) File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 81, in conv_carry_causal_3d out = op(x) File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl return self._call_impl(*args, **kwargs) ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl return forward_call(*args, **kwargs) File "H:\ROCm\ComfyUI\comfy\ops.py", line 516, in forward return self.forward_comfy_cast_weights(*args, **kwargs) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ File "H:\ROCm\ComfyUI\comfy\ops.py", line 508, in forward_comfy_cast_weights weight, bias, offload_stream = cast_bias_weight(self, input, offloadable=True) ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ROCm\ComfyUI\comfy\ops.py", line 302, in cast_bias_weight offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking) File "H:\ROCm\ComfyUI\comfy\ops.py", line 163, in cast_modules_with_vbar xfer_dest = get_cast_buffer(dest_size) File "H:\ROCm\ComfyUI\comfy\ops.py", line 122, in get_cast_buffer cast_buffer = comfy.model_management.get_aimdo_cast_buffer(offload_stream, device) File "H:\ROCm\ComfyUI\comfy\model_management.py", line 1221, in get_aimdo_cast_buffer cast_buffer = comfy_aimdo.vram_buffer.VRAMBuffer(DEFAULT_AIMDO_CAST_BUFFER_RESERVATION_SIZE, device.index) File "H:\ROCm\comfy-aimdo\comfy_aimdo\vram_buffer.py", line 34, in __init__ raise RuntimeError("VRAM reservation failed") RuntimeError: VRAM reservation failed ``` which looks similar to CUDA's https://github.com/Comfy-Org/comfy-aimdo/issues/49.
bsakaguchi commented 2026-10-26 22:23:37 +00:00
Author
Owner
Copy link

@alexsarmiento commented on GitHub (May 13, 2026):

access violation reading 0x00000000000000E0 is a rocm internal issue, addressed in ROCm/rocm-systems#6051.

After this fix I got:

0 models unloaded.
Model AutoencoderKL prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached.
aimdo: src/vrambuf.c:42:ERROR:vrambuf_create: 0 16777216k
!!! Exception during processing !!! VRAM reservation failed
Traceback (most recent call last):
  File "H:\ROCm\ComfyUI\execution.py", line 535, in execute
    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, v3_data=v3_data)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\execution.py", line 335, in get_output_data
    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, v3_data=v3_data)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\execution.py", line 309, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "H:\ROCm\ComfyUI\execution.py", line 297, in process_inputs
    result = f(**inputs)
  File "H:\ROCm\ComfyUI\nodes.py", line 316, in decode
    images = vae.decode(latent)
  File "H:\ROCm\ComfyUI\comfy\sd.py", line 1014, in decode
    model_management.raise_non_oom(e)
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
  File "H:\ROCm\ComfyUI\comfy\model_management.py", line 286, in raise_non_oom
    raise e
  File "H:\ROCm\ComfyUI\comfy\sd.py", line 1007, in decode
    out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True)
          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\comfy\ldm\models\autoencoder.py", line 254, in decode
    dec = self.decoder(dec, **decoder_kwargs)
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl
    return forward_call(*args, **kwargs)
  File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 808, in forward
    h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs)
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl
    return forward_call(*args, **kwargs)
  File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 224, in forward
    h = conv_carry_causal_3d(h, self.conv2, conv_carry_in=conv_carry_in, conv_carry_out=conv_carry_out)
  File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 81, in conv_carry_causal_3d
    out = op(x)
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl
    return forward_call(*args, **kwargs)
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 516, in forward
    return self.forward_comfy_cast_weights(*args, **kwargs)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 508, in forward_comfy_cast_weights
    weight, bias, offload_stream = cast_bias_weight(self, input, offloadable=True)
                                   ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 302, in cast_bias_weight
    offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking)
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 163, in cast_modules_with_vbar
    xfer_dest = get_cast_buffer(dest_size)
  File "H:\ROCm\ComfyUI\comfy\ops.py", line 122, in get_cast_buffer
    cast_buffer = comfy.model_management.get_aimdo_cast_buffer(offload_stream, device)
  File "H:\ROCm\ComfyUI\comfy\model_management.py", line 1221, in get_aimdo_cast_buffer
    cast_buffer = comfy_aimdo.vram_buffer.VRAMBuffer(DEFAULT_AIMDO_CAST_BUFFER_RESERVATION_SIZE, device.index)
  File "H:\ROCm\comfy-aimdo\comfy_aimdo\vram_buffer.py", line 34, in __init__
    raise RuntimeError("VRAM reservation failed")
RuntimeError: VRAM reservation failed

which looks similar to CUDA's #49.

Closing then since this is a rocm issue

<!-- gh-comment-id:4436498025 --> @alexsarmiento commented on GitHub (May 13, 2026): > `access violation reading 0x00000000000000E0` is a rocm internal issue, addressed in [ROCm/rocm-systems#6051](https://github.com/ROCm/rocm-systems/pull/6051). > > After this fix I got: > > ``` > 0 models unloaded. > Model AutoencoderKL prepared for dynamic VRAM loading. 159MB Staged. 0 patches attached. > aimdo: src/vrambuf.c:42:ERROR:vrambuf_create: 0 16777216k > !!! Exception during processing !!! VRAM reservation failed > Traceback (most recent call last): > File "H:\ROCm\ComfyUI\execution.py", line 535, in execute > 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, v3_data=v3_data) > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > File "H:\ROCm\ComfyUI\execution.py", line 335, in get_output_data > 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, v3_data=v3_data) > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > File "H:\ROCm\ComfyUI\execution.py", line 309, in _async_map_node_over_list > await process_inputs(input_dict, i) > File "H:\ROCm\ComfyUI\execution.py", line 297, in process_inputs > result = f(**inputs) > File "H:\ROCm\ComfyUI\nodes.py", line 316, in decode > images = vae.decode(latent) > File "H:\ROCm\ComfyUI\comfy\sd.py", line 1014, in decode > model_management.raise_non_oom(e) > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^ > File "H:\ROCm\ComfyUI\comfy\model_management.py", line 286, in raise_non_oom > raise e > File "H:\ROCm\ComfyUI\comfy\sd.py", line 1007, in decode > out = self.first_stage_model.decode(samples, **vae_options).to(device=self.output_device, dtype=self.vae_output_dtype(), copy=True) > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^ > File "H:\ROCm\ComfyUI\comfy\ldm\models\autoencoder.py", line 254, in decode > dec = self.decoder(dec, **decoder_kwargs) > File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl > return self._call_impl(*args, **kwargs) > ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ > File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl > return forward_call(*args, **kwargs) > File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 808, in forward > h1 = self.up[i_level].block[i_block](h1, temb, conv_carry_in, conv_carry_out, **kwargs) > File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl > return self._call_impl(*args, **kwargs) > ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ > File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl > return forward_call(*args, **kwargs) > File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 224, in forward > h = conv_carry_causal_3d(h, self.conv2, conv_carry_in=conv_carry_in, conv_carry_out=conv_carry_out) > File "H:\ROCm\ComfyUI\comfy\ldm\modules\diffusionmodules\model.py", line 81, in conv_carry_causal_3d > out = op(x) > File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1779, in _wrapped_call_impl > return self._call_impl(*args, **kwargs) > ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ > File "H:\ROCm\.venv\Lib\site-packages\torch\nn\modules\module.py", line 1790, in _call_impl > return forward_call(*args, **kwargs) > File "H:\ROCm\ComfyUI\comfy\ops.py", line 516, in forward > return self.forward_comfy_cast_weights(*args, **kwargs) > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^ > File "H:\ROCm\ComfyUI\comfy\ops.py", line 508, in forward_comfy_cast_weights > weight, bias, offload_stream = cast_bias_weight(self, input, offloadable=True) > ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > File "H:\ROCm\ComfyUI\comfy\ops.py", line 302, in cast_bias_weight > offload_stream = cast_modules_with_vbar([s], dtype, device, bias_dtype, non_blocking) > File "H:\ROCm\ComfyUI\comfy\ops.py", line 163, in cast_modules_with_vbar > xfer_dest = get_cast_buffer(dest_size) > File "H:\ROCm\ComfyUI\comfy\ops.py", line 122, in get_cast_buffer > cast_buffer = comfy.model_management.get_aimdo_cast_buffer(offload_stream, device) > File "H:\ROCm\ComfyUI\comfy\model_management.py", line 1221, in get_aimdo_cast_buffer > cast_buffer = comfy_aimdo.vram_buffer.VRAMBuffer(DEFAULT_AIMDO_CAST_BUFFER_RESERVATION_SIZE, device.index) > File "H:\ROCm\comfy-aimdo\comfy_aimdo\vram_buffer.py", line 34, in __init__ > raise RuntimeError("VRAM reservation failed") > RuntimeError: VRAM reservation failed > ``` > > which looks similar to CUDA's [#49](https://github.com/Comfy-Org/comfy-aimdo/issues/49). Closing then since this is a rocm issue
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
master
prs/cuda-not-ready-retry
prs/vbar-writeback
dev/windows-integrated-cuda
prs/mcdm-dxcore
dev/mcdm-dxcore
prs/null-guard-phys-free
dev/windows-integrated-ram
chore/node-runtime-refresh-2026-09-28
dev/alias-range-search-review
dev/disk-id
dev/non-graph-free
comfyanonymous-patch-1
prs/memory-compiler-abort
dev/malloc-graph-rogues
prs/memory-compiler
dev/malloc-graph-capture-finalize
dev/malloc-graph-stream-debug
prs/nvml-512
dev/nvml-pressure-headroom-768mb
fix/cla-api-failure-handling
dev/memory-graph-compiler-review-v2
dev/nvml-pressure-headroom-1gb
dev/nvml-pressure-headroom-512mb
dev/memory-graph-cuda-graphs
dev/memory-graph-scope-diagnostics
dev/cuda-capture-pressure-fix
dev/memory-graph-variable-calls
dev/memory-graph-linked-events
dev/competition-green
dev/competition-blue
dev/competition-yellow
dev/primary-load-default-stream
dev/threaded-loaded-errors
dev/remove-legacy-cla
dev/async-clamp
dev/nvml
dev/loop-graph-cache-series
dev/loop-malloc-compiler
dev/detail
dev/no-sequential-scan-benchmark
dev/threaded-loader-2
dev/protected-mem
dev/threaded-loader-3
alexisrolland-patch-2
alexisrolland-patch-1
dev/xpu-support
xpu-support
add-cla-workflow
dev/double-sync
dev/simple-miss-alloc
dev/mem-exp
dev/vbar-waterbug
dev/set-watermark
dev/multi-gpu-ring-split
dev/threaded-mmap
dev/direct-read
dev/mark-cold
dev/hostbuf-truncate-0-fix
dev/v4-cleanup
dev/sync-prewarm
dev/hostbuf-unreg
dev/dgx-spark
dev/unified-memory
stable-0.2
multigpu-thread-safety
multigpu-phase6
dev/213-general
dev/no-win-cudart
dev/vanilla-cuda-malloc
dev/memcpy
dev/threaded-gc
dev/lock-async-list
dev/vbar-virt-lead
dev/vbars-analyze
dev/less-throttle
dev/mallocs
prs/cleanup
dev/empty-cache
dev/torch-gc
dev/rattus/rbar
v0.6.0
v0.5.6
v0.5.5
v0.5.4
v0.5.3
v0.5.2
v0.5.1
v0.5.0
v0.4.15
v0.4.14
v0.4.13
v0.4.12
v0.4.11
v0.4.10
v0.4.9
v0.4.8
v0.4.7
v0.4.6
v0.4.5
v0.4.4
v0.4.3
v0.4.2
v0.4.1
v0.4.0
v0.3.0
v0.2.14
v0.0.214
v0.0.213
v0.2.12
v0.2.11
v0.2.10
v0.2.9
v0.2.8
v0.2.7
v0.2.6
v0.2.5
v0.2.4
v0.2.3
v0.2.2
v0.2.1
v0.2.0
v0.1.8
v0.1.7
v0.1.6
v0.1.5
v0.1.4
v0.1.3
v0.1.2
v0.1.1
v0.1.0
v0.0.4
v0.0.3
v0.0.2
v0.0.1
v0.0.0
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
Comfy-Org/comfy-aimdo#15
No description provided.