Skip to content

MOE dispatch_GEMM on Blackwell - #34

Draft
CalebZ9909 wants to merge 7 commits into
mainfrom
dispatch-gemm-blackwell
Draft

MOE dispatch_GEMM on Blackwell#34
CalebZ9909 wants to merge 7 commits into
mainfrom
dispatch-gemm-blackwell

Conversation

@CalebZ9909

@CalebZ9909 CalebZ9909 commented Jul 28, 2026

Copy link
Copy Markdown
Collaborator

This PR is for providing support of dispatch-GEMM on blackwell

@CalebZ9909

CalebZ9909 commented Jul 28, 2026

Copy link
Copy Markdown
Collaborator Author

@MaoZiming
I will focus on 1 pass workflow currently, dev plan

  1. Replace all RDMA with LSA
  2. Replace wgmma with tcgen05 (change of warp specialization)
  3. perf tuning (dynamic scheduling and GEMM-TMA-EPILOGUE overlapping)

@CalebZ9909
CalebZ9909 marked this pull request as draft July 30, 2026 08:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant