Hongwei Chen
Hongwei Chen
Hi @rpgmaker , can you shared the full error message you get to help us identify the cause? Thanks!
Can you rerun the script after setting the environment variable `CUDA_LAUNCH_BLOCKING=1` by ``` export CUDA_LAUNCH_BLOCKING=1 ``` and then share the full stack trace? Thanks!
It appears that `pin_memory()` leads to this error. Can you try `pin_memory: false` to see if it can solve this problem? Memory pinning can incur high memory consumption and cause...
Hi @tjruwase , can you have a look at it? Related PR: #4131
Hi @whlook , if you are using Nvidia GPU, can you run your code with CUDA sanitizer and share the output if any? ```TORCH_CUDA_SANITIZER=1 python your_code.py``` FYI: https://pytorch.org/docs/stable/cuda._sanitizer.html
I tried on the AMD zen4 CPU, and the code works OK. The `cpu_arch` function returns `-march=native`. I don't understand how the cpu_arch function returns `-march=x86-64-v3` instead of `-march=native`. Can...