完成项目1,cpu优化#48

Open

usersforsomebody wants to merge 27 commits intoInfiniTensor:mainfrom

usersforsomebody:main

Commits on Jan 22, 2026

Pass the test,may be need more review and check
usersforsomebody
committed

Commits on Jan 24, 2026

Finish the taks 2.1 Argmax
usersforsomebody
committed

Commits on Jan 25, 2026

Commits on Jan 26, 2026

Finish the Task-2.4 RMS
usersforsomebody
committed

Commits on Jan 27, 2026

Finish the task 2-5 RoPE and fix the kernel
usersforsomebody
committed

Commits on Jan 28, 2026

Finish The Task 2-6 attention is all your need
usersforsomebody
committed

Commits on Jan 29, 2026

Finish all the Task 1 Task 2
usersforsomebody
committed

Commits on Jan 31, 2026

Finsh the bind
usersforsomebody
committed

Commits on Feb 1, 2026

Finish forward init
usersforsomebody
committed

Commits on Feb 2, 2026

kv_cache
usersforsomebody
committed

Commits on Mar 8, 2026

perf: enable OpenMP and parallelize linear operator (4x-5x speedup)
usersforsomebody
committed

Commits on Mar 9, 2026

perf(ops): 分割 linear 循环块 (Tiling) 试图优化缓存命中，但因内部 utils::cast 阻断 SIMD 导致收益受限，作为引入 BLAS 库的性能参照
 usersforsomebody
committed

Commits on Mar 13, 2026

perf(linear): AVX2 SIMD 优化 f32/bf16/fp16 矩阵乘法，提取共享 SIMD 工具函数
 usersforsomebody
committed

Commits on Mar 14, 2026

Commits on Mar 15, 2026

feat(linear): 集成 OpenBLAS sgemm 加速 linear 算子
 usersforsomebody
committed

Commits on Mar 16, 2026