训练rtdetr模型时，报XDNN_RUNTIME_ERROR

[xiejx]

环境
系统：麒麟V10
xpu：昆仑k200

问题描述
训练rtdetr模型时，报如下的错误：

Traceback (most recent call last):
  File "tools/train.py", line 202, in <module>
    main()
  File "tools/train.py", line 198, in main
    run(FLAGS, cfg)
  File "tools/train.py", line 151, in run
    trainer.train(FLAGS.eval)
  File "/home/bml/storage/PaddleYOLO_20250225/ppdet/engine/trainer.py", line 566, in train
    loss.backward()
  File "/opt/conda/lib/python3.7/site-packages/decorator.py", line 232, in fun
    return caller(func, *(extras + args), **kw)
  File "/opt/conda/lib/python3.7/site-packages/paddle/fluid/wrapped_decorator.py", line 25, in __impl__
    return wrapped_func(*args, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/paddle/fluid/framework.py", line 602, in __impl__
    return func(*args, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/paddle/fluid/dygraph/varbase_patch_methods.py", line 323, in backward
    [scaled_loss], grad_tensor, retain_graph
OSError: (External) copy XDNN Error, XDNN_RUNTIME_ERROR  (at /data/paddle/Paddle/paddle/phi/kernels/xpu/elementwise_add_grad_kernel.cc:84)