mirror of
https://github.com/deepseek-ai/DeepEP.git
synced 2026-11-05 00:47:32 +00:00
[PR #287] [MERGED] Use TMA to optimize internode combine. #504
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#504
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/287
Author: @sphish
Created: 7/9/2025
Status: ✅ Merged
Merged: 7/10/2025
Merged by: @LyricZhao
Base:
main← Head:internode-combine-tma📝 Commits (10+)
8096742Let forwarders use a dedicated SM70edc81Shuffle rdma idxe94752aSender use TMA.499ec88Adjust the tuning chunk size.d6c3fe4Modify NVL chunk layout.bcab6b2Update some combine config.7449df8Small linte686dcaMerge remote-tracking branch 'origin/main' into internode-combine-tmaa4cbf5cMinor fix9212853Overlap TMA store📊 Changes
3 files changed (+94 additions, -72 deletions)
View changed files
📝
csrc/kernels/internode.cu(+90 -68)📝
deep_ep/buffer.py(+3 -3)📝
tests/test_internode.py(+1 -1)📄 Description
No description provided
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.