llama.cpp 发布 b12074：重构 delta_net 卷积状态
原标题：b12074
AI 摘要
llama.cpp 发布 b12074 版本，主要变更是对 delta_net_base 的卷积状态进行重构：改为通过视图访问卷积状态，修复 sm tensor，并将卷积窗口的回滚槽位索引改为在主机端填充的 I32 输入，替代原先在设备端计算的 arange 与 argsort，同时 meta 后端不再需要为 ARANGE 做镜像拆分。该版本同步提供 macOS、Linux、Windows、Android 等多平台及 CUDA、Vulkan、ROCm、SYCL、OpenVINO 等后端构建产物。
正文节选
<details open> delta_net_base: refactor conv state (#30878) * delta_net: access conv state via views * fix sm tensor * delta_net: fill the conv slot index on the host The rollback slot index of the conv windows is a constant per graph shape. It is now an I32 input of the recurrent memory filled on the host, instead of an arange and an argsort computed on the device, and the meta backend no longer needs a mirrored split for ARANGE. --------- Co-authored-by: Pascal <admin@serveurperso.com>