llama.cpp

ggml-cpu: Optimized risc-v cpu q1_0 dot

#22768

Merged