🐛 Describe the bug

Hello,

torch.export.export seems to produce invalid code for Tensor.split when used with meta device.

I am not sure if this is something that needs to be fixed, but here is the report anyway.

Please see the snippet below:

import torch


class MyModule(torch.nn.Module):
    def __init__(self):
        super().__init__()

    def forward(self, X):
        X.split(2, 1)


# This one works without any problems:
my_module = MyModule()
inp = torch.rand(10, 10)
exported = torch.export.export(my_module, (inp,))

# The generated code is:
# def forward(self, L_X_: torch.Tensor):
#     l_x_ = L_X_
#     split = l_x_.split(2, 1);
#     l_x_ = split = None
#     return ()

# This one crashes with the error message `AttributeError: module 'torch._tensor' has no attribute 'split'`:
with torch.device("meta"):
    my_module = MyModule()
    inp = torch.rand(10, 10)
    exported = torch.export.export(my_module, (inp,))

# The generated code is:
# def forward(self, L_X_: torch.Tensor):
#     l_x_ = L_X_
#     split = torch._tensor.split(l_x_, 2, 1);
#     l_x_ = split = None
#     return ()

If you get an error regarding circular imports, you might want to import import torch._dynamo as well.

Thank you.

Versions

PyTorch version: 2.7.0
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: macOS 15.4 (arm64)
GCC version: Could not collect
Clang version: 17.0.0 (clang-1700.0.13.3)
CMake version: version 4.0.1
Libc version: N/A

Python version: 3.13.3 (main, Apr 9 2025, 03:47:57) [Clang 20.1.0 ] (64-bit runtime)
Python platform: macOS-15.4-arm64-arm-64bit-Mach-O
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Apple M2 Max

Versions of relevant libraries:
[pip3] Could not collect
[conda] numpy 2.2.0 pypi_0 pypi

cc @chauhang @penguinwu @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @amjames @avikchaudhuri @gmagogsfm @zhxchen17 @tugsbayasgalan @angelayi @suo @ydwu4