Summary

When logical_or is called on tensors constructed from invalid string inputs, the API triggers a segmentation fault.

Code to reproduce bug

import oneflow as flow

# Create two tensors with correct data types
tensor1 = flow.tensor(["False", "True"], dtype=flow.bool)
tensor2 = flow.tensor(["True", "False"], dtype=flow.bool)
result = tensor1.logical_or(tensor2)

output:

Segmentation fault (core dumped)

System Information

• What is your OneFlow installation (pip, source, dockerhub): pip
• OS: Ubuntu 22.04.5 LTS
• OneFlow version (run python3 -m oneflow --doctor):
  path: ['/root/anaconda3/envs/oneflow1.0.0/lib/python3.10/site-packages/oneflow']
  version: 0.9.0
  git_commit: 381b12c
  cmake_build_type: Release
  rdma: True
  mlir: True
• Python version: 3.10.0
• CUDA driver version: 12.2
• GPU models: NVIDIA GeForce RTX 4090
• Other info: None