-
Notifications
You must be signed in to change notification settings - Fork 521
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
CUDA error if enabling compile_prefill for quantization model (int8) #137
Comments
Seeing similar issues with AMD gpus as well.
|
Observations:
So it seems to me the error is related to some interactions b/w the compiled prefill and decode kernels |
Looks like prefill compile can work, if I change |
Is there a resolution to this problem for
|
@griff4692 Does #137 (comment) work? |
Nope unfortunately -- it looks like in current code the |
@griff4692 It seems you hit a different issue other than this one, I tried your command and it works well at gpt-fast. So I suspect it's some change on the context-compression side triggered a cudagraph error. I'm looking at which triggers it now. |
The prefill issue w.r.t assertion (Nvidia) and memory fault (AMD) should be fixed by 2c33914 |
Repro command:
Errors:
generated kernel file: https://gist.github.com/yanboliang/6f5c1171e63909b995b5372dc7c88ab7
The text was updated successfully, but these errors were encountered: