Dsv4 sparse prefill optimization by skipping odd tail block#187
Draft
Njuapp wants to merge 5 commits into
Draft
Conversation
build: replace torch/extension.h with libtorch headers build: remove torch/python.h include from kerutils tensor helpers build: avoid utils.h include collision with sgl-kernel fix: keep FMHACutlassSM100FwdRun ABI and cast internally compat: convert sparse prefill max_logits/lse back to log2
Co-authored-by: Bruce Wu <mogicianwu@fb.com>
Njuapp
changed the title
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Optimize DS v4 prefill HCA kernel by skipping odd tail blocks.
For ISL=8k, we can cut down HCA kernel cost by 20%.