[CUDA] Update sm check for flash attention#24584
Merged
Conversation
hanbitmyths
approved these changes
Apr 29, 2025
baijumeswani
approved these changes
Apr 29, 2025
vraspar
pushed a commit
that referenced
this pull request
May 1, 2025
### Description Currently, flash attention is only enabled for sm8x and sm90. That means blackwell GPU will not use flash attention. This change is enable flash attention for sm > 90. Note that the flash attention implementation is not optimized for blackwell, but shall be able to run in blackwell GPU. Future works: * Integrate flash attn for hopper: https://github.com/Dao-AILab/flash-attention/tree/main/hopper * Integrate fmha for blackwell: https://github.com/NVIDIA/cutlass/tree/main/examples/77_blackwell_fmha * Update cudnn and cudnn frontend to latest version (so that we can use the cudnn flash attention for blackwell). ### Motivation and Context ORT GENAI is slow in RTX 5090
jywu-msft
pushed a commit
that referenced
this pull request
May 1, 2025
### Description Cherry pick the following into [rel-1.22.0](https://github.com/microsoft/onnxruntime/tree/rel-1.22.0) - (#24491) - (#24509) - (#24564) - (#24574) - (#24582) - (#24584) - (#24568) - (#24587) - (#24563) - (#24592) - (#24526) - (#24552) - (#24588) - (#24605) - (#24606) --------- Co-authored-by: Jing Fang <[email protected]> Co-authored-by: Tianlei Wu <[email protected]> Co-authored-by: Baiju Meswani <[email protected]> Co-authored-by: Scott McKay <[email protected]> Co-authored-by: Mark Schofield <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Edward Chen <[email protected]> Co-authored-by: Ashwath Shankarnarayan <[email protected]> Co-authored-by: saurabh <[email protected]> Co-authored-by: Adrian Lizarraga <[email protected]> Co-authored-by: Hector Li <[email protected]>
ankitm3k
pushed a commit
to intel/onnxruntime
that referenced
this pull request
May 12, 2025
### Description Currently, flash attention is only enabled for sm8x and sm90. That means blackwell GPU will not use flash attention. This change is enable flash attention for sm > 90. Note that the flash attention implementation is not optimized for blackwell, but shall be able to run in blackwell GPU. Future works: * Integrate flash attn for hopper: https://github.com/Dao-AILab/flash-attention/tree/main/hopper * Integrate fmha for blackwell: https://github.com/NVIDIA/cutlass/tree/main/examples/77_blackwell_fmha * Update cudnn and cudnn frontend to latest version (so that we can use the cudnn flash attention for blackwell). ### Motivation and Context ORT GENAI is slow in RTX 5090
Contributor
|
This PR has been included in the |
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
4 participants
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.
Description
Currently, flash attention is only enabled for sm8x and sm90. That means blackwell GPU will not use flash attention. This change is enable flash attention for sm > 90.
Note that the flash attention implementation is not optimized for blackwell, but shall be able to run in blackwell GPU.
Future works:
Motivation and Context
ORT GENAI is slow in RTX 5090