Fix memory layout of attention outputs #2582
Open
Conversation
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
Contributor
There was a problem hiding this comment.
Pull request overview
This PR fixes the memory layout of attention outputs in the XPU implementation by adding contiguous permutation operations and replacing empty placeholder tensors with properly initialized values.
Key Changes:
- Added contiguous memory layout conversion for attention output tensor
- Replaced empty placeholder tensors with initialized tensors using
at::fullandat::scalar_tensor
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
CuiYifeng
reviewed
Dec 23, 2025
Comment on lines
+371
to
+372
| Tensor out = | ||
| attention.permute({0, 2, 1, 3}).contiguous().permute({0, 2, 1, 3}); |
Contributor
There was a problem hiding this comment.
Why does out Tensor need to be contiguous in BLHE format?
Comment on lines
+375
to
+378
| at::full( | ||
| {B, H, (compute_log_sumexp ? ceil_div(L, kAlignLSE) * kAlignLSE : 0)}, | ||
| 0.0, | ||
| attention.options()), |
Contributor
There was a problem hiding this comment.
Why do we need to align L to 32?
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Outdated suggestions cannot be applied.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.
Fix memory layout of attention outputs