Skip to content
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.

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

stack smashing detected in begin_forward when compiling directly from the repo #166

Open
mkrima opened this issue Mar 8, 2024 · 1 comment

Comments

@mkrima
Copy link

mkrima commented Mar 8, 2024

Hello,

When I compile flashinfer directly from the repo, running begin_forward of BatchPrefillWithPagedKVCacheWrapper crashes with the error:

*** stack smashing detected ***: terminated

Running the same code with the version installed from pip works without a problem. Any suggestions what could be the problem?

Thank you.

@Qubitium
Copy link
Contributor

Qubitium commented Mar 10, 2024

@mkrima Please provide more env info such as os, cuda, torch version.

I believe the pip wheels are compiled va this ci code for cuda 11-8/12.1 and torch 2.1/2.2:

https://github.com/flashinfer-ai/flashinfer/blob/main/scripts/run-ci-build-wheel.sh

I would recommend not deviating too much from the min cuda 11.8 and min torch 2.1 requirements if you are doing custom compile.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants