Update non_xla attention to properly support paged_attention dynamo code path#7022
Merged
Update non_xla attention to properly support paged_attention dynamo code path#7022
Conversation
JackCaoG
approved these changes
May 2, 2024
alanwaketan
approved these changes
May 3, 2024
| attn_output = attn_weight @ v | ||
| return attn_output | ||
| # Return orignal shape of q. | ||
| return torch.empty_like(q) |
Collaborator
There was a problem hiding this comment.
Do you know if this actually initialize anything? I hope not.
Collaborator
There was a problem hiding this comment.
yea it is worth checking out, in above we have a warning about the q should be on meta device, I think running ops on meta_tensor will not allocate any device memory.
Collaborator
Author
There was a problem hiding this comment.
According to https://pytorch.org/docs/stable/generated/torch.empty_like.html, it seems like emtpy_like returns an uninitialized tensor. Along with the meta tensor check above, I think we should be good.
Collaborator
Author
|
Thanks for the reviews, I'll go ahead and merge this as the CIs (including TPU CI) are all green. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Test plan:
+ TPU CI