返回全部动态

llama.cpp b12188:修复 server 父子 slot 调度顺序

原标题:b12188

llama.cpp Releases一手来源开源质量 64

AI 摘要

llama.cpp 发布 b12188 版本,主要修复 server 端多 slot 调度问题:当子任务落在比父 slot 更低的 id 上时,子 slot 会先于父 slot 被访问,导致多等一个 batch 才处理后缀;现在改为先访问非子 slot 再访问活跃子 slot,使被父 slot 唤醒的子 slot 能在同一 batch 加入。同时将 systemone 测试改为通过 __TEST_TAG_WAIT_PARENT_SLOT__、__TEST_TAG_PROMPT_CACHE_RESTORE__ 等测试标签匹配事件,避免依赖自由格式调试信息。该版本同步提供 macOS、Linux、Windows、Android 等多平台预编译包。

以上摘要由 AI 生成,可能存在误差。事实请以原文为准。

正文节选

<details open> server: visit parent slots before child slots (#30779) * server: visit parent slots before child slots A parent slot wakes its waiting child slots once it finishes the shared prompt prefix, while slots are visited in id order. When a child task lands on a lower slot id than its parent, the child is visited before the parent wakes it and waits an extra batch to process its suffix. Non-child slots are now visited before active child slots, so every child woken by its parent joins


发布时间:2026-11-07 19:58
抓取时间:2026-11-07 20:14
来源机构:ggml-org
阅读原文github.com