[PR #195] [MERGED] Add low-latency kernel PCIe usage flag #458

New issue

Closed

opened 2026-10-26 10:00:59 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 10:00:59 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/DeepEP/pull/195
Author: @LyricZhao
Created: 6/9/2025
Status: ✅ Merged
Merged: 6/9/2025
Merged by: @LyricZhao

Base: main ← Head: pcie-flag

📝 Commits (2)

71b5140 Add low-latency kernel usage flag
d860dbd Update comments

📊 Changes

6 files changed (+57 additions, -13 deletions)

View changed files

📝 csrc/deep_ep.cpp (+15 -2)
📝 csrc/deep_ep.hpp (+6 -0)
📝 csrc/kernels/api.cuh (+4 -3)
📝 csrc/kernels/internode_ll.cu (+18 -8)
📝 deep_ep/buffer.py (+13 -0)
📝 tests/test_low_latency.py (+1 -0)

📄 Description

No description provided

_{🔄 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/195 **Author:** [@LyricZhao](https://github.com/LyricZhao) **Created:** 6/9/2025 **Status:** ✅ Merged **Merged:** 6/9/2025 **Merged by:** [@LyricZhao](https://github.com/LyricZhao) **Base:** `main` ← **Head:** `pcie-flag` --- ### 📝 Commits (2) - [`71b5140`](https://github.com/deepseek-ai/DeepEP/commit/71b5140c193ecaf07e82e7a551f26e0ae865ab47) Add low-latency kernel usage flag - [`d860dbd`](https://github.com/deepseek-ai/DeepEP/commit/d860dbd9ab209a019bdc6d33515f0358e090649c) Update comments ### 📊 Changes **6 files changed** (+57 additions, -13 deletions) <details> <summary>View changed files</summary> 📝 `csrc/deep_ep.cpp` (+15 -2) 📝 `csrc/deep_ep.hpp` (+6 -0) 📝 `csrc/kernels/api.cuh` (+4 -3) 📝 `csrc/kernels/internode_ll.cu` (+18 -8) 📝 `deep_ep/buffer.py` (+13 -0) 📝 `tests/test_low_latency.py` (+1 -0) </details> ### 📄 Description _No description provided_ --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-26 10:00:59 +00:00