VRAM memory Leak while using GPEN-BFR-512 model

[yorkane]

Easily to reproduce,
Aflter 24 images upscaledm throw out execeptions logs below:

ComfyUI Error Report

Error Details

• Node Type: easy forLoopEnd
• Exception Type: AttributeError
• Exception Message: 'NoneType' object has no attribute 'get_tensor_shape'

Stack Trace

  File "/home/aigc/comfyui/execution.py", line 323, in execute
    output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)

  File "/home/aigc/comfyui/execution.py", line 198, in get_output_data
    return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)

  File "/home/aigc/comfyui/custom_nodes/ComfyUI-0246/utils.py", line 353, in new_func
    res_value = old_func(*final_args, **kwargs)

  File "/home/aigc/comfyui/execution.py", line 169, in _map_node_over_list
    process_inputs(input_dict, i)

  File "/home/aigc/comfyui/execution.py", line 158, in process_inputs
    results.append(getattr(obj, func)(**inputs))

  File "/home/aigc/comfyui/custom_nodes/ComfyUI-Facerestore-Tensorrt/__init__.py", line 93, in main
    self.engine.allocate_buffers()

  File "/home/aigc/comfyui/custom_nodes/ComfyUI-Facerestore-Tensorrt/trt_utilities.py", line 229, in allocate_buffers
    shape = self.context.get_tensor_shape(name)

System Information

• ComfyUI Version: v0.2.6-1-g09fdb2b
• Arguments: main.py --listen --port 8192 --cuda-device 1
• OS: posix
• Python Version: 3.10.12 (main, Jul 29 2024, 16:56:48) [GCC 11.4.0]
• Embedded Python: false
• PyTorch Version: 2.4.1+cu121

Devices

• Name: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
  • Type: cuda
  • VRAM Total: 25386352640
  • VRAM Free: 305111178
  • Torch VRAM Total: 2382364672
  • Torch VRAM Free: 8560778