Modify notifydispatch to support DEEPEP_NORMAL_LONG_SEQ_ROUND up to 128.

[WSEmma]

Modify the notifydispatch file to support DEEPEP_NORMAL_LONG_SEQ_ROUND up to 128.

DEEPEP_NORMAL_LONG_SEQ_ROUND=128 DEEPEP_NORMAL_LONG_SEQ_PER_ROUND_TOKENS=512 verified.

[gemini-code-assist]

Summary of Changes

Hello @WSEmma, 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 significantly refactors the NotifyDispatch class to enhance its scalability by implementing a batch-based processing mechanism for handling a greater number of operational rounds. The changes enable the system to manage larger sequence lengths more efficiently, particularly for DEEPEP_NORMAL_LONG_SEQ_ROUND up to 128, by breaking down computations into smaller, manageable batches.

Highlights

• Increased Round Support: The notifydispatch logic has been updated to support DEEPEP_NORMAL_LONG_SEQ_ROUND values up to 128, enabling processing of longer sequences.
• Batch Processing for Rounds: A new BATCH_ROUND constant (set to 32) has been introduced, and core processing functions now iterate over rounds in batches. This refactoring likely improves memory efficiency and performance for larger round counts.
• Dynamic Buffer Initialization: Buffer initialization for recvDataBuf, recvCountBuf, and sendOffsetBuf has been moved from the constructor into the respective Build functions, allowing for more dynamic allocation based on batch processing.
• Renamed Variables: Variables related to 'send count' (sendCountBuf, sendCountTensor) have been renamed to 'receive count' (recvCountBuf, recvCountTensor), clarifying their purpose within the data flow.

🧠 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. ↩

[gemini-code-assist]

Code Review

The pull request refactors the NotifyDispatch class to introduce batch processing for DEEPEP_NORMAL_LONG_SEQ_ROUND up to 128. This involves modifying several Build* and Reorder* functions to iterate over batches, dynamically initializing buffers, and adjusting data copy and synchronization calls accordingly. Key changes include the introduction of a BATCH_ROUND constant, renaming sendCountBuf and sendCountTensor to recvCountBuf and recvCountTensor for better clarity, and adapting the logic to handle data processing in chunks.