CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.

[cvirxsc]

The program always stops at the same iteration and then reports an error. What should I do?

[meituan-gengyifei]

Thanks for your attention! We are trying to reproduce your error. Before that, you also can try this solution: https://blog.csdn.net/weixin_35848967/article/details/120200055?spm=1001.2101.3001.6650.1&utm_medium=distribute.pc_relevant.none-task-blog-2%7Edefault%7ECTRLIST%7Edefault-1-120200055-blog-123645660.pc_relevant_multi_platform_whitelistv1_exp2&depth_1-utm_source=distribute.pc_relevant.none-task-blog-2%7Edefault%7ECTRLIST%7Edefault-1-120200055-blog-123645660.pc_relevant_multi_platform_whitelistv1_exp2&utm_relevant_index=2

Might be able to pinpoint your problem more clearly