Segmentation fault occur when using index_copy_ #20322
Description
🐛 Bug
When using the index_copy_ function, the programme will throw a segmentation fault.
To Reproduce
Steps to reproduce the behavior:
Run the code below:
import torch
a = torch.randn(3, 5)
c = torch.zeros(3)
a.index_copy_(dim=1, index=torch.tensor([3]), source=c)MrFive@mrfive-home:~$ python
Python 3.7.2 (default, Dec 29 2018, 06:19:36)
[GCC 7.3.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> a = torch.randn(3, 5)
>>> b = torch.zeros(3, 4)
>>> c = torch.zeros(3)
>>> a.index_copy_(dim=1, index=torch.tensor([3]), source=c)
Segmentation fault (core dumped)I know the code following will work, but it seems that the issue above is an unexpected performent.
a.index_copy_(dim=1, index=torch.tensor([3]), source=c.unsqueeze(1))Environment
PyTorch version: 1.0.1.post2
Is debug build: No
CUDA used to build PyTorch: None
OS: Ubuntu 18.04.2 LTS
GCC version: (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0
CMake version: version 3.10.2
Python version: 3.7
Is CUDA available: No
CUDA runtime version: No CUDA
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
Versions of relevant libraries:
[pip3] numpy==1.16.2
[conda] blas 1.0 mkl https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
[conda] mkl 2019.1 144 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
[conda] mkl_fft 1.0.10 py37ha843d7b_0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
[conda] mkl_random 1.0.2 py37hd81dba3_0 https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
[conda] pytorch-cpu 1.0.1 py3.7_cpu_2 pytorch
[conda] torchvision-cpu 0.2.2 py_3 pytorch