mirror of
https://github.com/deepseek-ai/DeepEP.git
synced 2026-11-05 00:47:32 +00:00
[PR #410] [MERGED] Canonicalize TMA usages #555
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/DeepEP#555
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
📋 Pull Request Information
Original PR: https://github.com/deepseek-ai/DeepEP/pull/410
Author: @LyricZhao
Created: 9/16/2025
Status: ✅ Merged
Merged: 9/16/2025
Merged by: @LyricZhao
Base:
main← Head:elect-one-sync📝 Commits (8)
691803aRemove redundant TMA flushesff5ca40Less barrier initialization overhead0ba50a3Simplifyelect_one_sync97d3b41Useelect_one_syncinstead of lanesf73b88dMinor fix2bd032cPolish testing printse408d1eRefactor for internode kernels2cd957eBetter performance📊 Changes
6 files changed (+140 additions, -140 deletions)
View changed files
📝
csrc/kernels/internode.cu(+49 -44)📝
csrc/kernels/internode_ll.cu(+9 -14)📝
csrc/kernels/intranode.cu(+51 -59)📝
csrc/kernels/utils.cuh(+16 -17)📝
tests/test_internode.py(+13 -4)📝
tests/test_intranode.py(+2 -2)📄 Description
No description provided
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.