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

[GH-ISSUE #43] Concurrent read/write in internode notify_dispatch rdma channel prefix sum calculation #42

New issue
Closed
opened 2026-10-26 09:53:08 +00:00 by bsakaguchi · 2 comments
bsakaguchi commented 2026-10-26 09:53:08 +00:00
Owner
Copy link

Originally created by @wenduobaba on GitHub (Mar 5, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/43

Code

github.com/deepseek-ai/DeepEP@592296cd45/csrc/kernels/internode.cu (L387-L392)

Question

Hi team,

The above code reads to me that multiple threads could modify the same data in rdma_channel_prefix_matrix. Could you clarify the intended behavior?

Thanks!

Originally created by @wenduobaba on GitHub (Mar 5, 2025). Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/43 ### Code https://github.com/deepseek-ai/DeepEP/blob/592296cd45f458f7d4e2c7a370e1c755376cde3b/csrc/kernels/internode.cu#L387-L392 ### Question Hi team, The above code reads to me that multiple threads could modify the same data in `rdma_channel_prefix_matrix`. Could you clarify the intended behavior? Thanks!
bsakaguchi commented 2026-10-26 09:53:09 +00:00
Author
Owner
Copy link

@LyricZhao commented on GitHub (Mar 5, 2025):

You are right, seems a bug here. It happens to be correct because only one warp is doing that. I will fix it today. Thanks!

<!-- gh-comment-id:2699990473 --> @LyricZhao commented on GitHub (Mar 5, 2025): You are right, seems a bug here. It happens to be correct because only one warp is doing that. I will fix it today. Thanks!
bsakaguchi commented 2026-10-26 09:53:09 +00:00
Author
Owner
Copy link

@LyricZhao commented on GitHub (Mar 5, 2025):

Fixed in github.com/deepseek-ai/DeepEP@680e424bdc. Close it now, please feel free to reopen the issue if you have other questions.

<!-- gh-comment-id:2699996197 --> @LyricZhao commented on GitHub (Mar 5, 2025): Fixed in https://github.com/deepseek-ai/DeepEP/commit/680e424bdcfce4fc1351328f7d7a08687c79f250. Close it now, please feel free to reopen the issue if you have other questions.
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#42
No description provided.