New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

BatchPrefillWithPagedKVCacheSM90Run failed with error: operation not supported #807

Open

sfc-gh-yewang opened this issue Feb 12, 2025 · 0 comments · May be fixed by #808

Open

BatchPrefillWithPagedKVCacheSM90Run failed with error: operation not supported #807

sfc-gh-yewang opened this issue Feb 12, 2025 · 0 comments · May be fixed by #808

sfc-gh-yewang commented Feb 12, 2025

Env:
Linux + H100 + CUDA 12.4

To reproduce:
pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cu124
pip install flashinfer-python -i https://flashinfer.ai/whl/cu124/torch2.5

and run the example here:
https://github.com/flashinfer-ai/flashinfer/blob/main/flashinfer/prefill.py#L904-L985

got:
RuntimeError: BatchPrefillWithPagedKVCacheSM90Run failed with error: operation not supported

sfc-gh-yewang linked a pull request

that will close this issue

fix: Pass backend in BatchPrefillWith*KVCacheWrapper.plan() #808

Open

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment