[Fix]: add missing device attribute to ChunkCache

[leavelet]

Motivation

Fix #11492 by adding device attribute to 'ChunkCache'.

Modifications

'ChunkCache' will inherit device from its token_to_kv_pool_allocator.

Accuracy Tests

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.

[gemini-code-assist]

Summary of Changes

Hello @leavelet, 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 critical issue by adding a missing device attribute to the ChunkCache class. This ensures that the cache correctly identifies and operates on the appropriate hardware device, either by inheriting it from its associated allocator or defaulting to the CPU. This change is essential for maintaining proper device management and preventing potential errors related to device-specific memory operations within the system.

Highlights

• Device Attribute Addition: The ChunkCache class now includes a device attribute to explicitly track its operating device.
• Device Inheritance: The device attribute is initialized by inheriting the device from its token_to_kv_pool_allocator.
• CPU Fallback: If the token_to_kv_pool_allocator is not available, the device attribute defaults to torch.device("cpu").

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. ↩

[leavelet]

@cctry please review

[gemini-code-assist]

Code Review

This pull request adds a missing device attribute to the ChunkCache class. The implementation is sound, but I've identified a potential issue regarding the nullability of token_to_kv_pool_allocator. The current code defensively checks if this allocator exists, but other parts of the class do not, which could lead to runtime errors. My review comment suggests simplifying the code to align with the existing type hints, which would also resolve this inconsistency.

[gemini-code-assist]

This logic implies that token_to_kv_pool_allocator can be None. However, the type hint in the __init__ signature (BaseTokenToKVPoolAllocator) suggests it's non-nullable. This is inconsistent.

If token_to_kv_pool_allocator can be None, then cache_finished_req will raise an AttributeError at line 59, as it accesses self.token_to_kv_pool_allocator.free(...) without a check. Please ensure all usages of token_to_kv_pool_allocator are safe if it can be None, and consider updating the type hint to Optional[BaseTokenToKVPoolAllocator].

If token_to_kv_pool_allocator is never None (as the type hint suggests), this conditional logic is unnecessary and can be simplified.

        self.device = self.token_to_kv_pool_allocator.device

[cctry]

Sorry to bring in this bug. I am wondering when token_to_kv_pool_allocator will be None

[leavelet]

Sorry to bring in this bug. I am wondering when token_to_kv_pool_allocator will be None

Should we just change it to self.device = self.token_to_kv_pool_allocator.device? It should never be None anyway. Sorry about the mistake

[hzh0425]

hi @leavelet ， Could you help add a CI to cover this issue? It needs --disable-radix-cache and page_size > 1.