[Bugfix] minor fix mem device

[whybeyoung]

fix error caused by #11313

[gemini-code-assist]

Summary of Changes

Hello @whybeyoung, 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 implements a minor but crucial fix related to device assignment within the memory caching mechanism. It resolves an error that emerged from a prior change, ensuring that newly instantiated tensors are consistently allocated to the correct processing device. This correction is vital for maintaining the stability and efficiency of memory operations, preventing potential runtime issues caused by incorrect device placement.

Highlights

• Device Assignment Correction: Corrected the device parameter used when creating new tensors in the alloc_for_extend function, switching from batch.tree_cache.device to batch.device.
• Error Resolution: This change specifically addresses and fixes an error introduced by a previous pull request (Separate allocation logic from scheduler #11313), ensuring proper memory device handling.

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 error caused by a device mismatch during paged allocation. The fix correctly changes the device for tensor creation to batch.device, ensuring all tensors in the list comprehension are on the same device before concatenation. This resolves the crash and also improves robustness by not relying on the device attribute of tree_cache, which may not be present in all implementations. The change is correct and well-targeted.

[whybeyoung]

duplicate with #11493.
close .