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 #537] Add runtime warning for known num_tokens=512 bug #620

New issue
Open
opened 2026-10-27 21:04:35 +00:00 by bsakaguchi · 0 comments
bsakaguchi commented 2026-10-27 21:04:35 +00:00
Owner
Copy link

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/DeepEP/pull/537
Author: @yurekami
Created: 12/25/2025
Status: 🔄 Open

Base: main ← Head: warn-num-tokens-512

📝 Commits (2)

  • a5096e5 Add NVSHMEM guards for low_latency mask buffer functions
  • f71f4a5 Add runtime warning for known num_tokens=512 bug

📊 Changes

2 files changed (+21 additions, -0 deletions)

View changed files

📝 csrc/deep_ep.cpp (+12 -0)
📝 tests/test_low_latency.py (+9 -0)

📄 Description

Summary

  • Adds a runtime warning when users specify --num-tokens 512 in test_low_latency.py
  • The TODO comment at line 315 mentions this is a known bug
  • Warning alerts users before they encounter unexpected behavior

Changes

  • Added warning import and check after argument parsing
  • Warning message suggests alternative values (128, 256, or 1024)

Test plan

  • Run python tests/test_low_latency.py --num-tokens 512 and verify warning is displayed
  • Run with other num_tokens values to confirm no warning appears

🤖 Generated with Claude Code

🔄 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/537 **Author:** [@yurekami](https://github.com/yurekami) **Created:** 12/25/2025 **Status:** 🔄 Open **Base:** `main` ← **Head:** `warn-num-tokens-512` --- ### 📝 Commits (2) - [`a5096e5`](https://github.com/deepseek-ai/DeepEP/commit/a5096e53922553c3de2cbfdb38260a35c92291d3) Add NVSHMEM guards for low_latency mask buffer functions - [`f71f4a5`](https://github.com/deepseek-ai/DeepEP/commit/f71f4a571248364f6723aaad3efeca289bb45f8e) Add runtime warning for known num_tokens=512 bug ### 📊 Changes **2 files changed** (+21 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `csrc/deep_ep.cpp` (+12 -0) 📝 `tests/test_low_latency.py` (+9 -0) </details> ### 📄 Description ## Summary - Adds a runtime warning when users specify `--num-tokens 512` in `test_low_latency.py` - The TODO comment at line 315 mentions this is a known bug - Warning alerts users before they encounter unexpected behavior ## Changes - Added warning import and check after argument parsing - Warning message suggests alternative values (128, 256, or 1024) ## Test plan - [ ] Run `python tests/test_low_latency.py --num-tokens 512` and verify warning is displayed - [ ] Run with other num_tokens values to confirm no warning appears 🤖 Generated with [Claude Code](https://claude.com/claude-code) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
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#620
No description provided.