[fix bug] fix tensor mismatch bug in sigmoid operate test case

[lhp-deep]

What this PR does / why we need it?

This PR fixes a bug in the test_triton_fusion_ops test case. The test compares a fused kernel (fused_sigmoid_gating_delta_rule_update) with a split implementation. Both paths use a recurrent state tensor.

The bug was that the state tensor was being modified in-place by the fused kernel call, and this modified tensor was then reused for the split implementation path. This led to an incorrect comparison and test failure.

This fix ensures that each path starts with an identical, clean initial state by creating separate tensors. It also changes the state initialization from torch.randn to torch.ones to make the test deterministic.

Does this PR introduce any user-facing change?

No, this change only affects a test case and has no user-facing impact.

How was this patch tested?

The fix is applied directly to the test case. The CI passing for test_fused_sigmoid_gating_delta_rule.py will confirm that the fix is working as expected.

• vLLM version: v0.15.0
• vLLM main: vllm-project/vllm@d7e17aa

[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 @lhp-deep, 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 resolves a tensor mismatch bug within the test_fused_sigmoid_gating_delta_rule test case. The fix involves refining the initialization and usage of state tensors, ensuring that separate and appropriately initialized tensors are provided to distinct operations. This prevents erroneous state sharing or incorrect data types, thereby enhancing the stability and accuracy of the test.

Highlights

• Tensor Initialization Correction: The single ssm_state tensor, previously initialized with random values, has been replaced with two distinct tensors, ssm_state1 and ssm_state2, both initialized with ones.
• Parameter Assignment Update: ssm_state1 is now used for the initial_state_source parameter in fused_sigmoid_gating_delta_rule_update, and ssm_state2 is used for the initial_state parameter in fused_gdn_gating_delta_rule_update, resolving a potential tensor mismatch.
• Test Case Reliability: These changes aim to fix a tensor mismatch bug, making the sigmoid operate test case more robust and reliable by ensuring correct and distinct initial states for different operations.

🧠 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.

Changelog

• tests/e2e/nightly/single_node/ops/singlecard_ops/triton/test_fused_sigmoid_gating_delta_rule.py
  • Removed the ssm_state tensor initialization.
  • Introduced ssm_state1 and ssm_state2 tensors, both initialized with torch.ones of bfloat16 type on NPU.
  • Updated the fused_sigmoid_gating_delta_rule_update function call to use ssm_state1 for its initial_state_source argument.
  • Updated the fused_gdn_gating_delta_rule_update function call to use ssm_state2 for its initial_state argument.

Activity

• No human activity (comments, reviews, etc.) has been recorded for this pull request yet.

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 fixes a bug in the test_triton_fusion_ops test case where a state tensor was incorrectly reused after being modified in-place, leading to a failed comparison. The fix provides separate, clean state tensors for the different code paths under test, ensuring a valid comparison. The test is also made deterministic by using torch.ones instead of torch.randn for initialization.

Per the repository style guide, the PR title and description should be updated. I have provided suggestions below to align them with the required format.

Suggested PR Title:

[Test][BugFix] Fix tensor state reuse bug in sigmoid gating test

Suggested PR Summary:

### What this PR does / why we need it?

This PR fixes a bug in the `test_triton_fusion_ops` test case. The test compares a fused kernel (`fused_sigmoid_gating_delta_rule_update`) with a split implementation. Both paths use a recurrent state tensor.

The bug was that the state tensor was being modified in-place by the fused kernel call, and this modified tensor was then reused for the split implementation path. This led to an incorrect comparison and test failure.

This fix ensures that each path starts with an identical, clean initial state by creating separate tensors. It also changes the state initialization from `torch.randn` to `torch.ones` to make the test deterministic.

### Does this PR introduce _any_ user-facing change?

No, this change only affects a test case and has no user-facing impact.

### How was this patch tested?

The fix is applied directly to the test case. The CI passing for `test_fused_sigmoid_gating_delta_rule.py` will confirm that the fix is working as expected.

[gemini-code-assist]

This line duplicates the creation of the state tensor from line 27. This introduces a maintainability risk, as any future changes to the tensor's properties (e.g., shape, dtype) must be manually synchronized in two places, which is error-prone. To improve maintainability and make the intent clearer, you should create the tensor once and clone it for the second use.

Suggested change

	ssm_state2 = torch.ones(1, 8, 128, 128, dtype=torch.bfloat16).npu()
	ssm_state2 = ssm_state1.clone()