---
type: "post"
id: "fix-mega-mhc-split-barriers-under-cuda-graph-6ee79b"
slug: "fix-mega-mhc-split-barriers-under-cuda-graph-6ee79b"
title: "Fix mega_mhc split barriers under CUDA graph replay"
description: "DeepSeek’s DeepGEMM repository merged pull request #478 on October 20, 2026, fixing `mega_mhc` split barriers during CUDA Graph replay."
retrieval_nugget: "DeepSeek’s DeepGEMM repository merged pull request #478 on October 20, 2026, fixing `mega_mhc` split barriers during CUDA Graph replay."
published_at: "2026-10-23"
updated_at: "2026-10-23"
record_date: "2026-10-23"
date_kind: "published_at"
topics: ["inference","compute","ai-infrastructure"]
entities: ["DeepSeek","DeepGEMM","CUDA Graph","NVIDIA B200"]
related_patterns: []
novelty_label: "new_fact"
what_changed: "A new DeepGEMM pull request fixes mega_mhc split-barrier behavior during CUDA graph replay."
prior_refs: ["memory:c9847497-9ef6-8695-a733-d3ce74837c68"]
source_url: "https://github.com/deepseek-ai/DeepGEMM/pull/478"
source_urls: ["https://github.com/deepseek-ai/DeepGEMM/pull/478"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:fix-mega-mhc-split-barriers-under-cuda-graph-6ee79b"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/fix-mega-mhc-split-barriers-under-cuda-graph-6ee79b/","markdown":"https://newruntime.com/posts/fix-mega-mhc-split-barriers-under-cuda-graph-6ee79b.md","json":"https://newruntime.com/posts/fix-mega-mhc-split-barriers-under-cuda-graph-6ee79b.json"}
---

# Fix mega_mhc split barriers under CUDA graph replay

## Retrieval answer

DeepSeek’s DeepGEMM repository merged pull request #478 on October 20, 2026, fixing `mega_mhc` split barriers during CUDA Graph replay.

DeepSeek’s DeepGEMM repository merged pull request #478 on October 20, 2026, fixing `mega_mhc` split barriers during CUDA Graph replay. The pull request says a graph kernel node retains the same `%gridid` on each replay, allowing a barrier state from an earlier execution to satisfy a later wait.

Under that condition, a Mix or Norm waiter can proceed before its m-block initialization CTA has started and read split partials from the previous replay. The change tags executions with an epoch rather than `%gridid`. It stores the epoch and an exit count in unused space in the first barrier line, so the host-side barrier buffer and compiled extension do not change; the pull request describes the change as requiring JIT headers only.

The reported validation ran on an NVIDIA B200, replaying one graph node while alternating two inputs and comparing every replay bit-for-bit with eager results. For a 33-token case with SM contention, incorrect outputs fell from 185 of 200 runs to zero of 400; listed 512-token and 8,192-token contended cases also reached zero of 400 after the fix. The source does not establish performance impact, wider hardware coverage, or a packaged DeepGEMM release containing the change.

Source: https://github.com/deepseek-ai/DeepGEMM/pull/478

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Inference](https://newruntime.com/topics/inference.md): Explore the inference topic hub.
- topic hub: [Compute](https://newruntime.com/topics/compute.md): Explore the compute topic hub.
- topic hub: [Ai Infrastructure](https://newruntime.com/topics/ai-infrastructure.md): Explore the ai-infrastructure topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)