Skip to content

Fix the issue of HCCL buffer tiling verification failure during one round of testing.#280

Merged
Yael-X merged 1 commit intosgl-project:mainfrom
oagniqgnat:fix_1round_hccl
Dec 25, 2025
Merged

Fix the issue of HCCL buffer tiling verification failure during one round of testing.#280
Yael-X merged 1 commit intosgl-project:mainfrom
oagniqgnat:fix_1round_hccl

Conversation

@oagniqgnat
Copy link
Copy Markdown
Contributor

@oagniqgnat oagniqgnat commented Dec 25, 2025

This PR fixes the issue where the tiling verification fails due to not running the ant moving task, bs=1800, and specifying HCCL_BUFFSIZE=1536.

@gemini-code-assist
Copy link
Copy Markdown
Contributor

Summary of Changes

Hello @oagniqgnat, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses a critical issue causing HCCL buffer tiling verification failures during testing. The primary change involves refining the buffer size calculation in cam_moe_combine_normal_tiling.cc to accurately account for the actual batch size processed in a round, preventing an incorrect buffer allocation. Additionally, a minor, redundant print statement was introduced in a Python test file.

Highlights

  • HCCL Buffer Size Calculation Fix: Corrected the calculation of actualSize for HCCL buffers by introducing realMaxBs and using std::min(perRoundTokens, realMaxBs) to determine realBs, resolving a buffer tiling verification failure.
  • Logging Update: Updated the error log message in OP_LOGE to reflect the use of realBs in the buffer size calculation, providing more accurate debugging information.
  • Redundant Test Output: A duplicate print(f"rank {rank} PASSED") statement was added in test_intranode.py.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request addresses an HCCL buffer tiling verification failure by correctly calculating the required buffer size using the minimum of perRoundTokens and realMaxBs. This is a good fix. My review includes a minor correction for a misleading log message in the C++ code and points out a duplicated line in a Python test file that should be removed.

@oagniqgnat oagniqgnat force-pushed the fix_1round_hccl branch 2 times, most recently from 4505317 to 3d4339e Compare December 25, 2025 08:12
@Yael-X Yael-X merged commit ebf10b3 into sgl-project:main Dec 25, 2025
4 checks passed
oagniqgnat added a commit to oagniqgnat/sgl-kernel-npu that referenced this pull request Dec 27, 2025
* upstream/main:
  modify split_qkv_rmsnorm_rope (sgl-project#282)
  bump version to 2025.12.25 (sgl-project#281)
  l2 norm const parameter change (sgl-project#276)
  Fix the issue of HCCL buffer tiling verification failure during one round of testing. (sgl-project#280)
AndyKong2020 pushed a commit to AndyKong2020/sgl-kernel-npu that referenced this pull request Mar 24, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants