[Bug] ComfyUI中使用 Model Graph Saver 节点不能保存正确的graph

[LingXuanYin]

Your current environment information

Collecting environment information... PyTorch version: 2.1.0+cu121 Is debug build: False CUDA used to build PyTorch: 12.1 ROCM used to build PyTorch: N/A

OneFlow version: path: ['/root/miniconda3/lib/python3.10/site-packages/oneflow'], version: 0.9.1.dev20240701+cu121, git_commit: ec7b682, cmake_build_type: Release, rdma: True, mlir: True, enterprise: False Nexfort version: none OneDiff version: unknown version OneDiffX version: none

OS: Ubuntu 22.04.1 LTS (x86_64) GCC version: (Ubuntu 11.3.0-1ubuntu1~22.04) 11.3.0 Clang version: Could not collect CMake version: version 3.27.7 Libc version: glibc-2.35

Python version: 3.10.8 (main, Nov 24 2022, 14:13:03) [GCC 11.2.0] (64-bit runtime) Python platform: Linux-5.15.0-83-generic-x86_64-with-glibc2.35 Is CUDA available: True CUDA runtime version: 11.8.89 CUDA_MODULE_LOADING set to: LAZY GPU models and configuration: GPU 0: NVIDIA GeForce RTX 4090 Nvidia driver version: 535.104.05 cuDNN version: Probably one of the following: /usr/lib/x86_64-linux-gnu/libcudnn.so.8.6.0 /usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.6.0 /usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.6.0 /usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.6.0 /usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.6.0 /usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.6.0 /usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.6.0 HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True

CPU: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 46 bits physical, 57 bits virtual Byte Order: Little Endian CPU(s): 128 On-line CPU(s) list: 0-127 Vendor ID: GenuineIntel Model name: Intel(R) Xeon(R) Platinum 8352V CPU @ 2.10GHz CPU family: 6 Model: 106 Thread(s) per core: 2 Core(s) per socket: 32 Socket(s): 2 Stepping: 6 CPU max MHz: 3500.0000 CPU min MHz: 800.0000 BogoMIPS: 4000.00 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 invpcid_single ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect wbnoinvd dtherm ida arat pln pts avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid fsrm md_clear pconfig flush_l1d arch_capabilities Virtualization: VT-x L1d cache: 3 MiB (64 instances) L1i cache: 2 MiB (64 instances) L2 cache: 80 MiB (64 instances) L3 cache: 108 MiB (2 instances) NUMA node(s): 2 NUMA node0 CPU(s): 0-31,64-95 NUMA node1 CPU(s): 32-63,96-127 Vulnerability Gather data sampling: Mitigation; Microcode Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable Vulnerability Retbleed: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected

Versions of relevant libraries: [pip3] diffusers==0.27.2 [pip3] numpy==1.24.4 [pip3] onnx==1.16.1 [pip3] onnxruntime==1.18.0 [pip3] onnxruntime-gpu==1.18.0 [pip3] open-clip-torch==2.23.0 [pip3] torch==2.1.0 [pip3] torchaudio==2.1.0+cu121 [pip3] torchsde==0.2.6 [pip3] torchvision==0.16.0+cu121 [pip3] transformers==4.42.3 [pip3] triton==2.1.0 [conda] numpy 1.24.4 pypi_0 pypi [conda] open-clip-torch 2.23.0 pypi_0 pypi [conda] torch 2.1.0 pypi_0 pypi [conda] torchaudio 2.1.0+cu121 pypi_0 pypi [conda] torchsde 0.2.6 pypi_0 pypi [conda] torchvision 0.16.0+cu121 pypi_0 pypi [conda] triton 2.1.0 pypi_0 pypi

🐛 Describe the bug

0. 在ComfyUI中使用Load Checkpoint - OneDiff节点加速一个带IPA的自定义的工作流，运行，发现IPA无法生效。删除graph文件并重启comfyUI，更换工作流为ipadapter_advanced.json并使用上述onediff节点进行加速，运行，发现IPA生效。对比两次生成的graph文件，IPA生效时生成的graph文件大小为不生效时的两倍，但其文件名一致。

1. 删除graph文件并重启comfyUI，使用onediff加速的工作流ipadapter_advanced.json，运行，IPA生效，删除graph文件（此graph文件为较大的），不重启comfyUI，使用onediff加速上述我自定义的工作流，在KSampler节点后添加了节点Model Graph Saver，其中输入samplers来自KSampler节点，输入model与KSampler节点的输入相同，运行，报错如下

Warning: ModelGraphSaver will be deleted. Please use it with caution.
Graph saving...
!!! Exception during processing!!! Can't pickle <class 'onediff.infer_compiler.backends.oneflow.transform.builtin_transform.<class 'register_comfy.CrossAttentionPatch.Attn2Replace'>'>: attribute lookup <class 'register_comfy.CrossAttentionPatch.Attn2Replace'> on onediff.infer_compiler.backends.oneflow.transform.builtin_transform failed
Traceback (most recent call last):
  File "/root/ComfyUI/execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
  File "/root/ComfyUI/execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
  File "/root/ComfyUI/execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
  File "/root/ComfyUI/custom_nodes/onediff_comfy_nodes/extras_nodes/nodes_oneflow_booster.py", line 524, in save_graph
    save_graph(diffusion_model, filename_prefix, "cuda", subfolder="unet")
  File "/root/ComfyUI/custom_nodes/onediff_comfy_nodes/modules/oneflow/utils/__init__.py", line 64, in save_graph
    deploy_module.save_graph(graph_filename)
  File "/root/onediff/src/onediff/infer_compiler/backends/oneflow/deployable_module.py", line 212, in save_graph
    self.get_graph().save_graph(file_path, process_state_dict=process_state_dict)
  File "/root/onediff/src/onediff/infer_compiler/backends/oneflow/utils/cost_util.py", line 49, in clocked
    return func(*args, **kwargs)
  File "/root/onediff/src/onediff/infer_compiler/backends/oneflow/graph.py", line 65, in save_graph
    flow.save(state_dict, file_path)
  File "/root/miniconda3/lib/python3.10/site-packages/oneflow/framework/check_point_v2.py", line 720, in save
    pickled_bytes = pickle.dumps(obj)
_pickle.PicklingError: Can't pickle <class 'onediff.infer_compiler.backends.oneflow.transform.builtin_transform.<class 'register_comfy.CrossAttentionPatch.Attn2Replace'>'>: attribute lookup <class 'register_comfy.CrossAttentionPatch.Attn2Replace'> on onediff.infer_compiler.backends.oneflow.transform.builtin_transform failed

Prompt executed in 89.25 seconds

注意：此时依然生成了一个graph文件，但其大小是上述较大的graph文件的一半，且此次运行结果显示IPA未生效

2. 删除graph并重启comfyUI，重复1. 但不加入节点Model Graph Saver且不删除较大的graph文件，运行，IPA生效，再次运行，IPA生效

3. 1. 和2. 的所有控制台输出。上述所有graph文件具有相同的文件名SDXL_7ecb88_d49b38b9.graph

got prompt
model_type EPS
Using xformers attention in VAE
Using xformers attention in VAE
Cache lookup: Key='b303a9a0-6293-4280-86af-aaf69abe240c', Cached Model Type='<class 'NoneType'>'
Requested to load CLIPVisionModelProjection
Loading 1 new model
Requested to load SDXLClipModel
Loading 1 new model
Requested to load SDXL
Loading 1 new model
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 30/30 [01:08<00:00,  2.29s/it]
Requested to load AutoencoderKL
Loading 1 new model
Prompt executed in 113.55 seconds


got prompt
model_type EPS
Using xformers attention in VAE
Using xformers attention in VAE
Cache lookup: Key='fdb5fad2-2eef-4905-822f-82d40b08e6fc', Cached Model Type='<class 'NoneType'>'
Requested to load CLIPVisionModelProjection
Loading 1 new model
CLIP: [1girl, Seraphina Noir,

Brown almond-shaped eyes, Long black flowing hair, Fair skin, Blue strapless dress with earrings.]
Requested to load SDXLClipModel
Loading 1 new model
Requested to load SDXL
Loading 1 new model
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 24/24 [00:48<00:00,  2.04s/it]
Requested to load AutoencoderKL
Loading 1 new model
WAS Node Suite: Image file saved to: /root/ComfyUI/output/ComfyUI_0001.jpeg
Warning: ModelGraphSaver will be deleted. Please use it with caution.
Graph saving...
!!! Exception during processing!!! Can't pickle <class 'onediff.infer_compiler.backends.oneflow.transform.builtin_transform.<class 'register_comfy.CrossAttentionPatch.Attn2Replace'>'>: attribute lookup <class 'register_comfy.CrossAttentionPatch.Attn2Replace'> on onediff.infer_compiler.backends.oneflow.transform.builtin_transform failed
Traceback (most recent call last):
  File "/root/ComfyUI/execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
  File "/root/ComfyUI/execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
  File "/root/ComfyUI/execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
  File "/root/ComfyUI/custom_nodes/onediff_comfy_nodes/extras_nodes/nodes_oneflow_booster.py", line 524, in save_graph
    save_graph(diffusion_model, filename_prefix, "cuda", subfolder="unet")
  File "/root/ComfyUI/custom_nodes/onediff_comfy_nodes/modules/oneflow/utils/__init__.py", line 64, in save_graph
    deploy_module.save_graph(graph_filename)
  File "/root/onediff/src/onediff/infer_compiler/backends/oneflow/deployable_module.py", line 212, in save_graph
    self.get_graph().save_graph(file_path, process_state_dict=process_state_dict)
  File "/root/onediff/src/onediff/infer_compiler/backends/oneflow/utils/cost_util.py", line 49, in clocked
    return func(*args, **kwargs)
  File "/root/onediff/src/onediff/infer_compiler/backends/oneflow/graph.py", line 65, in save_graph
    flow.save(state_dict, file_path)
  File "/root/miniconda3/lib/python3.10/site-packages/oneflow/framework/check_point_v2.py", line 720, in save
    pickled_bytes = pickle.dumps(obj)
_pickle.PicklingError: Can't pickle <class 'onediff.infer_compiler.backends.oneflow.transform.builtin_transform.<class 'register_comfy.CrossAttentionPatch.Attn2Replace'>'>: attribute lookup <class 'register_comfy.CrossAttentionPatch.Attn2Replace'> on onediff.infer_compiler.backends.oneflow.transform.builtin_transform failed

Prompt executed in 93.72 seconds


got prompt
model_type EPS
Using xformers attention in VAE
Using xformers attention in VAE
Cache lookup: Key='f9098578-f38c-4fa8-b5fe-47cad9a44cd5', Cached Model Type='<class 'NoneType'>'
Requested to load CLIPVisionModelProjection
Loading 1 new model
Requested to load SDXLClipModel
Loading 1 new model
Requested to load SDXL
Loading 1 new model
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 30/30 [01:01<00:00,  2.06s/it]
Requested to load AutoencoderKL
Loading 1 new model
Prompt executed in 100.32 seconds


got prompt
model_type EPS
Using xformers attention in VAE
Using xformers attention in VAE
Cache lookup: Key='c75095ff-76ae-47a7-854d-51cd394529a6', Cached Model Type='<class 'NoneType'>'
Requested to load CLIPVisionModelProjection
Loading 1 new model
CLIP: [1girl, Seraphina Noir,

Brown almond-shaped eyes, Long black flowing hair, Fair skin, Blue strapless dress with earrings.]
Requested to load SDXLClipModel
Loading 1 new model
Requested to load SDXL
Loading 1 new model
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 24/24 [00:04<00:00,  5.57it/s]
Requested to load AutoencoderKL
Loading 1 new model
WAS Node Suite: Image file saved to: /root/ComfyUI/output/ComfyUI_0001.jpeg
Prompt executed in 43.03 seconds


got prompt
CLIP: [1girl, Seraphina Noir,

Brown almond-shaped eyes, Long black flowing hair, Fair skin, Blue strapless dress with earrings.]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 24/24 [00:02<00:00, 11.00it/s]
WAS Node Suite: Image file saved to: /root/ComfyUI/output/ComfyUI_0002.jpeg
Prompt executed in 3.22 seconds