【main】【bugfix】fix: restrict default MLAPO activation to Decode nodes only

[fems14]

What this PR does / why we need it?

There is an issue with the current default logic for MLAPO (MLA Policy Optimization). By design, MLAPO should only be enabled by default on Decode (D) nodes. However, in hybrid (collocated prefill and decode) scenarios, the strategy is erroneously activated during the Prefill stage.
This PR corrects the default behavior to ensure that MLAPO is exclusively enabled for the Decoding phase. This prevents unexpected policy interference during Prefill and ensures optimal performance in hybrid deployment environments.

Does this PR introduce any user-facing change?

How was this patch tested?

• vLLM version: v0.14.1
• vLLM main: vllm-project/vllm@dc917cc

[github-actions]

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

• A PR should do only one thing, smaller PRs enable faster reviews.
• Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
• Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

[gemini-code-assist]

Summary of Changes

Hello @fems14, 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 bug where DeepSeek models failed in a PD mixed environment. The fix primarily involves correcting a typo in the naming of the MLAPO (Multi-Layer Attention Parallelism Optimization) feature and refining the logic for when this feature is enabled, specifically ensuring it's not activated on KV producer instances during decode operations. These changes aim to stabilize the attention mechanism's behavior in complex inference setups.

Highlights

• Typo Correction: Corrected a pervasive typo from 'malpo' to 'mlapo' in import statements, variable assignments, and function definitions across vllm_ascend/attention/mla_v1.py and vllm_ascend/attention/utils.py.
• MLAPO Enablement Logic Refinement: Updated the enabling_mlapo function to include an additional condition, ensuring that MLAPO is only enabled for decode instances that are not also KV producers. This prevents potential issues in mixed prefill-decode (PD mixed) scenarios.

🧠 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

This pull request addresses a runtime failure for DeepSeek models in mixed prefill/decode scenarios. The changes consist of two parts:

1. A typo correction, renaming enabling_malpo to enabling_mlapo across the codebase for consistency.
2. A logic refinement in the enabling_mlapo function to ensure the MLAPO optimization is only enabled for pure decode instances (i.e., KV cache consumers that are not also producers).

The code changes are correct and effectively resolve the underlying issue.

On a separate note, the pull request title and description do not fully adhere to the repository's style guide.

• The title should use [] instead of 【】 and include a [Module] tag (e.g., [Attention]).
• The description sections ('What this PR does / why we need it?', 'How was this patch tested?') should be filled out to provide context for reviewers.
  Please update the title and description to align with the contribution guidelines.

[leo-pony]

LGTM