PR #18840: [NVIDIA] Support larger head dim for cudnn fmha #19828
+45
−18
Conversation
This file contains 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
e80b34f to
fb4946b
Compare
Imported from GitHub PR #18840 Since [cudnn v9.5.0](https://docs.nvidia.com/deeplearning/cudnn/latest/release-notes.html#cudnn-9-5-0), the larger head dim of 256 is supported. This PR enables this improvement. cc @Cjkkkk Copybara import of the project: -- 723af68 by kaixih <[email protected]>: Support larger head dim for cudnn fmha -- 5177dbd by kaixih <[email protected]>: Add unit test -- 9e8d8be by kaixih <[email protected]>: Formatting -- 5d93712 by kaixih <[email protected]>: Address comments -- 02777e1 by kaixih <[email protected]>: Separate tests -- e4cc8eb by kaixih <[email protected]>: Clang format Merging this change closes #18840 FUTURE_COPYBARA_INTEGRATE_REVIEW=#18840 from kaixih:cudnn_fmha_large_head_dim e4cc8eb PiperOrigin-RevId: 700218013
fb4946b to
2fb0cf1
Compare
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Outdated suggestions cannot be applied.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.
PR #18840: [NVIDIA] Support larger head dim for cudnn fmha
Imported from GitHub PR #18840
Since cudnn v9.5.0, the larger head dim of 256 is supported. This PR enables this improvement.
cc @Cjkkkk
Copybara import of the project:
--
723af68 by kaixih [email protected]:
Support larger head dim for cudnn fmha
--
5177dbd by kaixih [email protected]:
Add unit test
--
9e8d8be by kaixih [email protected]:
Formatting
--
5d93712 by kaixih [email protected]:
Address comments
--
02777e1 by kaixih [email protected]:
Separate tests
--
e4cc8eb by kaixih [email protected]:
Clang format
Merging this change closes #18840
FUTURE_COPYBARA_INTEGRATE_REVIEW=#18840 from kaixih:cudnn_fmha_large_head_dim e4cc8eb