Skip to content

Implement lhs_quant_pack_qsi8d32p_f32 using Intrinsics

John McLoughlin requested to merge john/lhs_quant_pack_qsi8d32p4x16sb_f32_neon into main

  • Added vectorized Advanced SIMD to improve performance

  • Implemented targetting mr 4, kr 16, sr 2 & bl 32

  • New files kai_lhs_quant_pack_qsi8d32p4x16sb_f32_neon.c & .h

Signed-off-by: johmcl01 john.mcloughlin@arm.com

Merge request reports