deepseek-ai/DeepEP
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepEP.git synced 2026-11-05 00:47:32 +00:00
Code Issues 308 Projects Packages Wiki Activity

[PR #225] [CLOSED] Refactor and support TMA for internode kernels #476

New issue
Closed
opened 2026-10-26 10:01:13 +00:00 by bsakaguchi · 0 comments
bsakaguchi commented 2026-10-26 10:01:13 +00:00
Owner
Copy link

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/DeepEP/pull/225
Author: @LyricZhao
Created: 6/19/2025
Status: ❌ Closed

Base: main ← Head: internode-tma

📝 Commits (10+)

📊 Changes

4 files changed (+234 additions, -176 deletions)

View changed files

📝 csrc/kernels/buffer.cuh (+27 -18)
📝 csrc/kernels/internode.cu (+205 -156)
📝 csrc/kernels/utils.cuh (+1 -1)
📝 tests/test_low_latency.py (+1 -1)

📄 Description

Still a draft, do not merge!

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/deepseek-ai/DeepEP/pull/225 **Author:** [@LyricZhao](https://github.com/LyricZhao) **Created:** 6/19/2025 **Status:** ❌ Closed **Base:** `main` ← **Head:** `internode-tma` --- ### 📝 Commits (10+) - [`2445327`](https://github.com/deepseek-ai/DeepEP/commit/24453275e34b6c214615db74df4b47da614b8094) Add `EP_TEST_LL_COMPATIBILITY` - [`3a3398f`](https://github.com/deepseek-ai/DeepEP/commit/3a3398f6864785b7ae6818ae775f662f6115ed8c) Minor fix - [`a0a6e22`](https://github.com/deepseek-ai/DeepEP/commit/a0a6e22eff25291ed57031428b0d1bf1c1cce10d) Fully remove forwarders' and NVL receivers' code - [`55bbd8c`](https://github.com/deepseek-ai/DeepEP/commit/55bbd8caafd7533290fb3979c137c5e530345613) Add impl - [`177e491`](https://github.com/deepseek-ai/DeepEP/commit/177e491e92fcb7fce89b5f352bc2b5f7ec1bb893) Fix send heads - [`49b9084`](https://github.com/deepseek-ai/DeepEP/commit/49b9084268d858399eea476fb825242d1140dca8) Fix several bugs - [`cd5c57f`](https://github.com/deepseek-ai/DeepEP/commit/cd5c57fb2af52655b3490797f0a985784d2cc8cf) Fix compilation - [`8da790e`](https://github.com/deepseek-ai/DeepEP/commit/8da790e3f3232f48838a47c5a69ecd8eaddad110) Fix the shifted buffer pointer - [`371df2d`](https://github.com/deepseek-ai/DeepEP/commit/371df2da52c815c80518741250ba667dc0f770a8) Fix bugs - [`74afd75`](https://github.com/deepseek-ai/DeepEP/commit/74afd75df26d0b60e701de6b303688f233184622) Fix bugs ### 📊 Changes **4 files changed** (+234 additions, -176 deletions) <details> <summary>View changed files</summary> 📝 `csrc/kernels/buffer.cuh` (+27 -18) 📝 `csrc/kernels/internode.cu` (+205 -156) 📝 `csrc/kernels/utils.cuh` (+1 -1) 📝 `tests/test_low_latency.py` (+1 -1) </details> ### 📄 Description Still a draft, do not merge! --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
bsakaguchi 2026-10-26 10:01:13 +00:00
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
hybrid-ep
chore/remove-clang-format-yapf
v2.5
nvDev
try-fix-cu128
mori-ep
antgroup-opt
compat-old-torch
add-linter
eager-rdma
v1.2.1
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepEP#476
No description provided.