[PD-Disagg] Unify prefill info data transition flow, all with PrefillServerInfo#19195
[PD-Disagg] Unify prefill info data transition flow, all with PrefillServerInfo#19195
PrefillServerInfo#19195Conversation
Summary of ChangesHello @hnyls2002, 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 refactors the handling of prefill parallelization information within the disaggregation module. By consolidating disparate data points into a single, well-defined dataclass, the change significantly improves the organization and maintainability of the codebase. This unification simplifies how parallelization parameters are passed, stored, and accessed throughout the system, leading to a more robust and easier-to-understand data flow for disaggregated prefill operations. Highlights
🧠 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
Activity
Using Gemini Code AssistThe 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
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 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
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
The pull request unifies the prefill information data transition flow by introducing the PrefillParallelInfo dataclass and replacing multiple individual tables with a single prefill_info_table. This refactoring improves code maintainability and readability. However, the current implementation of data fetching is less robust than the previous one, as it lacks explicit type enforcement and backward compatibility for JSON keys, which could lead to runtime errors during rolling updates or when encountering uninitialized values.
hnyls2002
changed the title
PrefillParallelInfoPrefillServerInfo
|
/rerun-stage stage-c-test-8-gpu-h20 |
|
✅ Triggered |
|
/rerun-stage stage-b-test-large-2-gpu |
|
✅ Triggered |
…o xverse_moe * 'xverse_moe' of https://github.com/xiaobaicxy/sglang: (275 commits) fix: add missing blank line after docstring in serving_transcription.py (sgl-project#19206) Whisper model support & `/v1/audio/transcriptions` endpoint & benchmark (sgl-project#16983) fix: patch docker image fixes (sgl-project#19100) [PD-Disagg] Unify prefill info data transition flow, all with `PrefillServerInfo` (sgl-project#19195) [CI] Tiny enhance the dp attention load blance benchmark (sgl-project#19194) add new ci user (sgl-project#19133) [CI] fix the teardown output of disaggregation test (sgl-project#19193) [PD-Disagg] Support query dp rank from bootstrap server. (sgl-project#19168) [Kernel Slimming] Migrate AWQ marlin repack kernel to JIT (sgl-project#18949) [Diffusion] Match rotary_embedding module name style (sgl-project#19179) [Refactor] Split rotary_embedding.py into a modular package (sgl-project#19144) [NPU] bump sgl-kernel-npu to 2026.02.01.post2 (sgl-project#19178) Use single mma warp group for short q_len in FA to optimize decoding performance (sgl-project#18985) Reorganize topk logic to clean up code and expose logical experts (sgl-project#16945) [ROCm] Use unreg path for custom all-reduce during CUDA graph capture (sgl-project#19162) [diffusion] feat: detect Flux2 custom VAE path from component_paths (sgl-project#19170) [AMD] ENV flags tuning and cleanup (sgl-project#19176) Fix bench_one_batch_server by moving the print statements (sgl-project#19175) Update rocm7.2 Dockerfile to install amdsmi for QuickReduce Initialization (sgl-project#19091) Revert "Refactor graph input buffers (sgl-project#18991)" (sgl-project#19173) ...
1 participant
Summary
Dicttables inCommonKVManagerwith a singlePrefillServerInfodataclass andprefill_info_tabledataclasses.asdict()/PrefillServerInfo(**data)for serialization end-to-end between bootstrap server and decode fetcher_fetch_prefill_server_infofromCommonKVReceivertoCommonKVManagerprefill_page_size_tableandfollow_bootstrap_room_tableon node failure in mooncake/nixl