load_state_dict_into_model is slow.

[Difers]

测试环境

System environment:
    sys.platform: linux
    Python: 3.10.12 (main, Feb  4 2025, 14:57:36) [GCC 11.4.0]
    CUDA available: True
    MUSA available: False
    numpy_random_seed: 1519529526
    GPU 0,1,2,3,4,5,6,7: NVIDIA H800
    CUDA_HOME: /usr/local/cuda-12.9
    NVCC: Cuda compilation tools, release 12.9, V12.9.41
    GCC: x86_64-linux-gnu-gcc (Ubuntu 11.4.0-1ubuntu1~22.04.2) 11.4.0
    PyTorch: 2.6.0+cu124
    PyTorch compiling details: PyTorch built with:
  - GCC 9.3
  - C++ Version: 201703
  - Intel(R) oneAPI Math Kernel Library Version 2024.2-Product Build 20240605 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v3.5.3 (Git Hash 66f0cb9eb66affd2da3bf5f8d897376f04aae6af)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - LAPACK is enabled (usually provided by MKL)
  - NNPACK is enabled
  - CPU capability usage: AVX512
  - CUDA Runtime 12.4
  - NVCC architecture flags: -gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_90,code=sm_90
  - CuDNN 90.9  (built against CUDA 12.9)
    - Built with CuDNN 90.1
  - Magma 2.6.1
  - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, COMMIT_SHA=2236df1770800ffea5697b11b0bb0d910b2e59e1, CUDA_VERSION=12.4, CUDNN_VERSION=9.1.0, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOROCTRACER -DLIBKINETO_NOXPUPTI=ON -DUSE_FBGEMM -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wsuggest-override -Wno-psabi -Wno-error=old-style-cast -Wno-missing-braces -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, TORCH_VERSION=2.6.0, USE_CUDA=ON, USE_CUDNN=ON, USE_CUSPARSELT=1, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_GLOO=ON, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=1, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, USE_ROCM_KERNEL_ASSERT=OFF, 

    TorchVision: 0.21.0+cu124
    OpenCV: 4.11.0
    MMEngine: 0.10.6

Runtime environment:
    launcher: pytorch
    randomness: {'seed': None, 'deterministic': False}
    cudnn_benchmark: False
    mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0}
    dist_cfg: {'backend': 'nccl'}
    seed: None
    deterministic: False
    Distributed launcher: pytorch
    Distributed training: True
    GPU number: 128

16机测试全量sft，在初次加载权重时耗时较长：
查看调用逻辑为_load_pretrained_model -> load_state_dict_into_model

DeepSeek-671B-SFT-Guide/code/xtuner/utils/handle_moe_load_and_save.py

Line 133 in ccf17c5

for name, param in module.named_parameters(

在这段处理中,会处理expert的参数以适应modeling组网，这段参数处理自测耗时估算为3小时，处理一层layer的expert参数约3分钟，共需处理58层；