pytorch - 💡(How to fix) Fix torch.compile changes output strides for nearest F.interpolate on non-contiguous input

🐛 Describe the bug

torch.compile with the Inductor backend changes the output layout/strides for F.interpolate(..., mode="nearest") on a non-contiguous input tensor.

Eager execution and compiled execution produce equal values, but the output strides differ. Eager returns a non-contiguous output with strides (420, 1, 84, 14, 2), while the compiled version returns a contiguous output with strides (420, 210, 42, 7, 1).

import torch
import torch.nn.functional as F

def f(x):
    return F.interpolate(x, size=(5, 6, 7), mode="nearest")

x = torch.randn(2, 5, 2, 3, 2).permute(0, 4, 1, 2, 3)

eager = f(x)
compiled = torch.compile(f, backend="inductor")(x)

print("input_shape:", tuple(x.shape))
print("input_stride:", x.stride())
print("eager_stride:", eager.stride())
print("compiled_stride:", compiled.stride())
print("eager_contiguous:", eager.is_contiguous())
print("compiled_contiguous:", compiled.is_contiguous())
print("values_equal:", torch.equal(eager, compiled))

if torch.equal(eager, compiled) and eager.stride() != compiled.stride():
    raise SystemExit(0)

raise SystemExit(1)

Error logs

input_shape: (2, 2, 5, 2, 3)
input_stride: (60, 1, 12, 6, 2)
eager_stride: (420, 1, 84, 14, 2)
compiled_stride: (420, 210, 42, 7, 1)
eager_contiguous: False
compiled_contiguous: True
values_equal: True

Versions

PyTorch version: 2.13.0.dev20260513+cpu Is debug build: False CUDA used to build PyTorch: Could not collect ROCM used to build PyTorch: N/A

OS: Ubuntu 24.04.3 LTS (x86_64) GCC version: (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0 Clang version: 18.1.3 (1ubuntu1) CMake version: Could not collect Libc version: glibc-2.39

Python version: 3.11.15 (main, Mar 11 2026, 17:20:07) [GCC 14.3.0] (64-bit runtime) Python platform: Linux-6.17.0-20-generic-x86_64-with-glibc2.39 Is CUDA available: False CUDA runtime version: Could not collect CUDA_MODULE_LOADING set to: N/A GPU models and configuration: GPU 0: NVIDIA RTX 6000 Ada Generation GPU 1: NVIDIA RTX 6000 Ada Generation

Nvidia driver version: 570.211.01 cuDNN version: Could not collect Is XPU available: False HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True Caching allocator config: N/A

Versions of relevant libraries: [pip3] numpy==2.4.4 [pip3] torch==2.13.0.dev20260513+cpu [conda] numpy 2.4.4 pypi_0 pypi [conda] torch 2.13.0.dev20260513+cpu pypi_0 pypi

cc @chauhang @penguinwu @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @aakhundov @coconutruben @jataylo