[Cute,Fwd,Sm90] Ceil div in paged kv manager to prevent size 0#2446
Open
imbr92 wants to merge 2 commits intoDao-AILab:mainfrom
Open
[Cute,Fwd,Sm90] Ceil div in paged kv manager to prevent size 0#2446imbr92 wants to merge 2 commits intoDao-AILab:mainfrom
imbr92 wants to merge 2 commits intoDao-AILab:mainfrom
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
When running the SM90 cute-dsl fwd with kv cache with gemma 3 4B, I got
ValueError: Expected size in shape to be strictly positive, but got 0since n_block_size is 80 (or 64 for the local layers) but num_threads = 128. Added a ceil div to prevent this and added a correctness test to the existing cute test file. Also enabledtest_flash_attn_paged_deepseekfor SM90 since paged kv is now supported on SM90.Tests:
Ran on H200.
test_flash_attn_paged_deepseek