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

!!! Exception during processing !!! cudaMallocAsync does not yet support checkPoolLiveAllocations. #142

Open
jasoncow007 opened this issue Nov 13, 2024 · 5 comments

Comments

@jasoncow007
Copy link

!!! Exception during processing !!! cudaMallocAsync does not yet support checkPoolLiveAllocations. If you need it, please file an issue describing your use case.

@kijai
Copy link
Owner

kijai commented Nov 13, 2024

In regards to what exactly?

@jasoncow007
Copy link
Author

In regards to what exactly?

image
in this case

@kijai
Copy link
Owner

kijai commented Nov 25, 2024

For cudagraphs you need to disable cuda malloc, you can do that by launching comfy with --disable-cuda-malloc

I find inductor is more compatible and really fast since torch 2.5.1 though.

@jasoncow007
Copy link
Author

after updating torch 2.5.1, I found that it compiling much more frequently than before

@jasoncow007
Copy link
Author

For cudagraphs you need to disable cuda malloc, you can do that by launching comfy with --disable-cuda-malloc

I find inductor is more compatible and really fast since torch 2.5.1 though.

you are running comfy on python3.10 or 3.11?

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