-
Notifications
You must be signed in to change notification settings - Fork 1.6k
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
TRITON_INTERPRET failing #4755
Comments
Can you attach your script? |
Hey Keren! I am created a draft pr in Works fine without |
It might be difficult for me to work with the example you're showing. I would appreciate if you could provide me a single complete script. |
gotcha let me see if I can find what file is breaking and reproduce from there |
Isolated the issue to the repo. https://github.com/triton-lang/kernels/blob/main/kernels/flash_attention.py |
Can you let me know what inputs you're using? |
I'm not able to reproduce the problem using the default test cases |
@Jokeren
|
What do you mean different behavior? For example, is it the case that triton w/ and w/o interpreter output different results? Or triton failed to compile but interpreter can yield results? |
@Jokeren |
@brisker I'm still confused. Which results are you referring to? Is it I only see very little differences. Is it a problem for you? w/o interpreter
w/ interpreter
|
@Jokeren with interpret:
without interpret:
|
I understand what you referring to now. However, I cannot reproduce the results using triton/main |
how to use triton/main?
|
Yes |
@Jokeren I am using RTX 4060 to run this, w/o interpret is quite fast, but with interpret the code seems to be very slow.
|
Right, the interpreter uses CPU |
Wanted to flag this quickly. I am running a script that usually works without fail and added
TRITON_INTERPRET=1
which now gives me this errorWhen I do not use the flag I have no such errors
The text was updated successfully, but these errors were encountered: