[Fix] Fix compile error by HAOCHENYE · Pull Request #4357 · hpcaitech/ColossalAI (original) (raw)

📌 Checklist before creating the PR

• I have created an issue for this PR for traceability
• The title follows the standard format: [doc/gemini/tensor/...]: A concise description
• I have added relevant tags if possible for us to better distinguish different PRs

🚨 Issue number

Link this PR to your issue with words like fixed to automatically close the linked issue upon merge

e.g. fixed #1234, closed #1234, resolved #1234

📝 What does this PR do?

torch.cuda.get_arch_list() here:

will get the available GPU arch for the current PyTorch lib. For example, the PyTorch 2.0 is built by the hopper arch GPU, it will return a list:

['sm_37', 'sm_50', 'sm_60', 'sm_70', 'sm_75', 'sm_80', 'sm_86', 'sm_90']

However, my GPU is an ampere arch GPU, return a list includes sm_90 here will cause an compile error like:

nvcc fatal : Unsupported gpu architecture 'compute_90'.

Therefore, I think the valid arch returned here should be >= sm60 but <= sm86, which is the max available arch for my GPU

💥 Checklist before requesting a review

• I have linked my PR to an issue (instruction)
• My issue clearly describes the problem/feature/proposal, with diagrams/charts/table/code if possible
• I have performed a self-review of my code
• I have added thorough tests.
• I have added docstrings for all the functions/methods I implemented

⭐️ Do you enjoy contributing to Colossal-AI?

• 🌝 Yes, I do.
• 🌚 No, I don't.

Tell us more if you don't enjoy contributing to Colossal-AI.