[CUDA] update doc for cuda contrib op GroupQueryAttention - #29173
Merged
Conversation
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
4 tasks
tianleiwu
force-pushed
the
tlwu/update_cuda_gqa_doc
branch
from
June 20, 2026 17:13
f3146a9 to
6ebf863
Compare
Contributor
There was a problem hiding this comment.
Pull request overview
This PR relocates the GroupQueryAttention contrib-op documentation from docs/contrib_ops/gqa.md to docs/contrib_ops/cuda/gqa.md, and expands it to focus on CUDA kernel behavior (dispatch, XQA, cuDNN SDPA, Flash/MEA/unfused), build options, and profiling/testing guidance.
Changes:
- Move the GQA documentation into the CUDA-specific docs subtree (
docs/contrib_ops/cuda/). - Add new sections covering input formats, backend selection controls (
sdpa_kernel+ env vars), profiling/benchmarking, fast build options, and known limitations/future work.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.
| File | Description |
|---|---|
| docs/contrib_ops/gqa.md | Removes the previous top-level GQA documentation file. |
| docs/contrib_ops/cuda/gqa.md | Adds the relocated/expanded CUDA-focused GQA documentation (needs relative-link fixes after the move). |
jiafatom
approved these changes
Jun 22, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This pull request moves the
docs/contrib_ops/gqa.mddocumentation file for theGroupQueryAttentioncontrib operator.The file contained detailed information about the operator's schema, CUDA kernel backends, quantization, attention sink, and related implementation details.